{"id": "7ae33abe73137c9bf3e381e7b9cd258595066c50888b5cb89bd1bc1636bee602", "sources": ["arxiv", "semantic_scholar"], "title": "Theoretical Foundations of Continual Learning via Drift-Plus-Penalty", "abstract": "In many real-world settings, data streams are nonstationary and arrive sequentially, requiring learning systems to adapt continuously without retraining from scratch. Continual learning (CL) addresses this challenge by incorporating new tasks while mitigating catastrophic forgetting, where learning new information degrades performance on previously acquired knowledge. We introduce a control-theoretic perspective on CL that explicitly regulates the evolution of forgetting, framing adaptation as a controlled process subject to long-term stability constraints. We focus on replay-based CL, where a finite memory buffer stores representative samples from prior tasks. We propose COntinual Learning with Drift-Plus-Penalty (COLD), a continual learning framework based on the Drift-Plus-Penalty (DPP) principle from stochastic optimization. To facilitate analysis, we also consider an oracle variant, COLD-ORACLE, as a reference benchmark. At each task, both methods minimize the current task loss while maintaining a virtual queue that tracks deviations from long-term stability on previously learned tasks, capturing the stability-plasticity trade-off as a regulated dynamical process. We establish stability and convergence guarantees that characterize this trade-off through a tunable control parameter. Experiments on standard benchmarks demonstrate that COLD consistently outperforms a broad range of state-of-the-art CL methods while providing competitive and controllable forgetting behavior through explicit regulation of stability and plasticity.", "authors": ["Nazreen Shah", "Govinda Arya", "Bharath B. N.", "Ranjitha Prasad"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2026-06-07", "url": "https://arxiv.org/abs/2606.08452", "pdf_url": "https://arxiv.org/pdf/2606.08452v1", "arxiv_id": "2606.08452", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.35}
{"id": "4850e5ea6ef87ce087f03ca4b0eabc6e257dbb0860acc720c50dd9907e56c299", "sources": ["arxiv", "semantic_scholar"], "title": "Evaluating the Impact of Task Granularity on Catastrophic Forgetting in Continual Learning", "abstract": "Catastrophic forgetting, the abrupt loss of previously acquired knowledge upon learning new information, remains the central challenge in Continual Learning. This project investigates whether the order in which a model learns information affects how well it retains knowledge. Specifically, we ask: does learning general categories first (like \"animals\" vs \"vehicles\") before learning specific classes (like \"dog\" vs \"cat\") reduce forgetting compared to learning all classes at once? We test three approaches on CIFAR-100: (1) Coarse-to-Fine: train on 2 super-classes, then expand to 10 specific sub-classes, (2) Fine-to-Coarse: train on 10 sub-classes, then group into 2 super-classes, and (3) Flat: train on all 10 classes from the start. We use Elastic Weight Consolidation (EWC) to prevent forgetting during transitions. Our hypothesis is that learning general patterns first creates a stable foundation that helps the model retain knowledge when learning more detailed distinctions. We evaluate using standard metrics (accuracy, precision, recall, F1) plus continual learning metrics like backward transfer and forgetting rates. This work could inform how we design learning sequences for real-world systems that need to learn incrementally.", "authors": ["Emre Alyamac", "Himanshu Janmeda", "Shashwat Krishna", "Yash Vijay"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2026-06-06", "url": "https://arxiv.org/abs/2606.08013", "pdf_url": "https://arxiv.org/pdf/2606.08013v1", "arxiv_id": "2606.08013", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.35}
{"id": "b18fd7f43dfeabcfae6fe146458efb7e0cf2cf3e90648c5719ae460e645c15ea", "sources": ["arxiv", "semantic_scholar"], "title": "Catastrophic Forgetting as Accessibility Collapse: A Three-Level Framework for Knowledge Persistence in Continual Learning", "abstract": "Catastrophic forgetting is commonly interpreted as the irreversible erasure of previously acquired knowledge during sequential learning. In this work, we investigate an alternative perspective: that forgetting may arise not from complete destruction of task representations but from a loss of accessibility to preserved information. We introduce a three-level framework separating knowledge storage, representation, and accessibility, and evaluate each component through a series of continual-learning experiments on sequential CIFAR-100 classification using ResNet-18. Our analysis combines checkpoint persistence, linear probing, representation geometry, classifier-reset recovery, and layer-wise recoverability experiments. We observe complete behavioral forgetting of earlier tasks, with task accuracy collapsing from 54.8% to 0%, while linear probe performance retains approximately 76% of the original representational information. Furthermore, retraining only the final classifier restores 75.7% of the original task performance without modifying the backbone network. Layer-wise analysis reveals that early and intermediate layers preserve highly recoverable task information despite severe degradation at later stages. Projection-energy and principal-angle analyses indicate that retained knowledge persists as distributed high-dimensional representations rather than through preservation of a small dominant subspace. These findings suggest that catastrophic forgetting is better characterized as an accessibility failure than complete representational erasure, and that substantial task-relevant information remains embedded within neural representations even after functional forgetting has occurred.", "authors": ["Ayushman Trivedi", "Bhavika Melwani"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2026-06-04", "url": "https://arxiv.org/abs/2606.06032", "pdf_url": "https://arxiv.org/pdf/2606.06032v1", "arxiv_id": "2606.06032", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.35}
{"id": "065489ca5e818afb74c9b2672e38d9cf7e575bb2e927e5945881d4a9f3916acb", "sources": ["arxiv", "semantic_scholar"], "title": "Spurious Correlation Learning in Preference Optimization: Mechanisms, Consequences, and Mitigation via Tie Training", "abstract": "Preference learning methods like Direct Preference Optimization (DPO) are known to induce reliance on spurious correlations, leading to sycophancy and length bias in today's language models and potentially severe goal misgeneralization in future systems. In this work, we provide a unified theoretical analysis of this phenomenon, characterizing the mechanisms of spurious learning, its consequences on deployment, and a provable mitigation strategy. Focusing on log-linear policies, we show that standard preference-learning objectives induce reliance on spurious features at the population level through two channels: mean spurious bias and causal-spurious correlation leakage. We then show that this reliance creates an irreducible vulnerability to distribution shift: more data from the same training distribution fails to reduce the model's dependence on spurious features. To address this, we propose tie training, a data augmentation strategy using ties (equal-utility preference pairs) to introduce data-driven regularization. We demonstrate that this approach selectively reduces spurious learning without degrading causal learning. Finally, we validate our theory on log-linear models and provide empirical evidence that both the spurious learning mechanisms and the benefits of tie training persist for neural networks and large language models.", "authors": ["Christian Moya", "Alex Semendinger", "Guang Lin", "Elliott Thornley"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2026-05-11", "url": "https://arxiv.org/abs/2605.11134", "pdf_url": "https://arxiv.org/pdf/2605.11134v2", "arxiv_id": "2605.11134", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Proceedings of the 43rd International Conference on Machine Learning, 2026, Seoul, South Korea", "quality_score": 0.55}
{"id": "f3e3c9c5a0c133ff30f1c3db1da3a1dc576515ff46b68e23dfa408bdaa665872", "sources": ["arxiv", "semantic_scholar"], "title": "Overcoming Catastrophic Forgetting in Visual Continual Learning with Reinforcement Fine-Tuning", "abstract": "Recent studies suggest that Reinforcement Fine-Tuning (RFT) is inherently more resilient to catastrophic forgetting than Supervised Fine-Tuning (SFT). However, whether RFT (e.g., GRPO) can effectively overcome forgetting in challenging visual continual learning settings, such as class-incremental learning (CIL) and domain-incremental learning (DIL), remains an open problem. Through a pilot study, we confirm that while RFT consistently outperforms SFT, it still suffers from non-negligible forgetting. We empirically trace this bottleneck to Trajectory-level Drift Agnosticism: among candidate rollouts achieving identical task rewards, the KL divergence from the preceding-task policy varies substantially, which strongly correlates with catastrophic forgetting across sequential tasks. Motivated by this insight, we propose Retention-aware Policy Optimization (RaPO), a simple yet effective RFT method that explicitly mitigates forgetting through trajectory-level reward shaping. Specifically, RaPO comprises two core components: (1) Retention Reward that converts trajectory-level distribution drift into a continuous reward signal, preferentially reinforcing knowledge-preserving rollouts within each group; (2) Cross-Task Advantage Normalization (CTAN), which maintains a persistent exponential moving average of reward statistics across task boundaries to stabilize the optimization progress during continual learning. Leveraging the free-form textual generalization of MLLMs, we comprehensively evaluate RaPO across five visual continual learning settings. Extensive experiments demonstrate that RaPO achieves leading performance, substantially reducing catastrophic forgetting while preserving strong plasticity. To the best of our knowledge, this work represents the first systematic exploration of RFT in visual continual learning, offering insights that we hope will inspire future research.", "authors": ["Meng Lou", "Hanzhong Guo", "Linwei Chen", "Yizhou Yu"], "categories": ["cs.CV", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2026-05-10", "url": "https://arxiv.org/abs/2605.09640", "pdf_url": "https://arxiv.org/pdf/2605.09640v1", "arxiv_id": "2605.09640", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.35}
{"id": "21894e3295e39db44486c6233961e068c650f3d81d53bb682b2e40142fbcc54e", "sources": ["arxiv", "semantic_scholar"], "title": "Path-Coupled Bellman Flows for Distributional Reinforcement Learning", "abstract": "Distributional reinforcement learning (DRL) models the full return distribution, but existing finite-support or quantile-based methods rely on projections, while recent flow-based approaches can suffer from \\emph{boundary mismatch} at the flow source or from \\emph{high-variance} bootstrapping when current and successor noises are independent. We propose Path-Coupled Bellman Flows (PCBF), a continuous-time DRL method that learns return distributions with flow matching using \\textbf{source-consistent Bellman-coupled paths}: the current path starts from the required base prior at $t{=}0$, reaches the Bellman target at $t{=}1$, and maintains a pathwise affine relation to the successor flow at intermediate times (without requiring time-$t$ marginals to satisfy a distributional Bellman fixed point for all $t$). PCBF couples current and successor return flows through shared base noise and uses a $λ$-parameterized control-variate target: $λ{=}0$ recovers an unbiased sample Bellman target, while $λ{>}0$ trades controlled bias for variance reduction. Experiments on analytically tractable MRPs, OGBench, and D4RL show improved distributional fidelity and training stability, and competitive offline RL performance.", "authors": ["Boyang Xu", "Qing Zou", "Siqin Yang", "Hao Yan"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2026-05-07", "url": "https://arxiv.org/abs/2605.08253", "pdf_url": "https://arxiv.org/pdf/2605.08253v2", "arxiv_id": "2605.08253", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Proceedings of the 43rd International Conference on Machine Learning, Seoul, South Korea. PMLR 306, 2026", "quality_score": 0.55}
{"id": "4b34aaae8dfc185ede3e357ba6f020d259bba6b31b636d112040f5d3a8516493", "sources": ["arxiv", "semantic_scholar"], "title": "Sequential Learning and Catastrophic Forgetting in Differentiable Resistor Networks", "abstract": "Differentiable physical networks provide a simple setting in which learning can be studied through the interaction between trainable parameters and physical equilibrium constraints. We investigate sequential learning in differentiable resistor networks governed by Kirchhoff's laws. Although individual input--output mappings can be learned by gradient-based adjustment of edge conductances, sequential training on conflicting tasks produces catastrophic forgetting. We show that forgetting is controlled by task conflict and by the degree of adaptation to the new task. Uniform anchoring and normalised gradient-weighted anchoring reduce forgetting only by increasing the final loss on the new task, giving a clear forgetting--adaptation trade-off. We also show that forgetting is associated with localised conductance changes on high-current edges, giving a physical interpretation as reconfiguration of dominant transport pathways. Broader random-task ensembles show that the strongest forgetting occurs when the second task reverses the output ordering imposed by the first task. Finally, comparisons across Erdős--Rényi, small-world, scale-free, and random-geometric graph ensembles show that topology changes the forgetting--adaptation balance. These results position differentiable resistor networks as compact, physically interpretable testbeds for studying continual learning in tunable matter.", "authors": ["Maniru Ibrahim"], "categories": ["cs.LG", "cond-mat.dis-nn", "physics.comp-ph"], "fields_of_study": ["Computer Science", "Physics"], "published_date": "2026-05-02", "url": "https://arxiv.org/abs/2605.01383", "pdf_url": "https://arxiv.org/pdf/2605.01383v1", "arxiv_id": "2605.01383", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.35}
{"id": "fe4ea4a8991e1b6e8e3d6551b34d8b952295e01c8948c3c5879fd7a201eda05d", "sources": ["arxiv", "semantic_scholar"], "title": "CI-CBM: Class-Incremental Concept Bottleneck Model for Interpretable Continual Learning", "abstract": "Catastrophic forgetting remains a fundamental challenge in continual learning, in which models often forget previous knowledge when fine-tuned on a new task. This issue is especially pronounced in class incremental learning (CIL), which is the most challenging setting in continual learning. Existing methods to address catastrophic forgetting often sacrifice either model interpretability or accuracy. To address this challenge, we introduce ClassIncremental Concept Bottleneck Model (CI-CBM), which leverage effective techniques, including concept regularization and pseudo-concept generation to maintain interpretable decision processes throughout incremental learning phases. Through extensive evaluation on seven datasets, CI-CBM achieves comparable performance to black-box models and outperforms previous interpretable approaches in CIL, with an average 36% accuracy gain. CICBM provides interpretable decisions on individual inputs and understandable global decision rules, as shown in our experiments, thereby demonstrating that human understandable concepts can be maintained during incremental learning without compromising model performance. Our approach is effective in both pretrained and non-pretrained scenarios; in the latter, the backbone is trained from scratch during the first learning phase. Code is publicly available at github.com/importAmir/CI-CBM.", "authors": ["Amirhosein Javadi", "Tuomas Oikarinen", "Tara Javidi", "Tsui-Wei Weng"], "categories": ["cs.LG", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2026-04-16", "url": "https://arxiv.org/abs/2604.14519", "pdf_url": "https://arxiv.org/pdf/2604.14519v1", "arxiv_id": "2604.14519", "doi": "10.48550/arXiv.2604.14519", "citation_count": 0, "influential_citation_count": 0, "has_code": true, "code_url": null, "venue": "Transactions on Machine Learning Research, 2026", "quality_score": 0.8429}
{"id": "19dfee6d24c8475c737f33849a8c8848a5e2a2c5ae71ad8a451650e14909411a", "sources": ["arxiv", "semantic_scholar"], "title": "SOLAR: A Self-Optimizing Open-Ended Autonomous Agent for Lifelong Learning and Continual Adaptation", "abstract": "Despite the remarkable success of large language models (LLMs), they still face bottlenecks while deploying in dynamic, real-world settings with primary challenges being concept drift and the high cost of gradient-based adaptation. Traditional fine-tuning (FT) struggles to adapt to non-stationary data streams without resulting in catastrophic for getting or requiring extensive manual data curation. To address these limitations within the streaming and continual learning paradigm, we propose the Self-Optimizing Lifelong Autonomous Reasoner (SOLAR) which is an open-ended autonomous agent that leverages parameter-level meta-learning to self-improve, treating model weights as an environment for exploration. It initiates the process by consolidating a strong prior over common-sense knowledge making it effective for transfer-learning. By utilizing a multi-level reinforcement learning approach, SOLAR autonomously discovers adaptation strategies, enabling efficient test-time adaptation to unseen domains. Crucially, SOLAR maintains an evolving knowledge base of valid modification strategies, implicitly acting as an episodic memory buffer to balance plasticity (adaptation to new tasks) and stability (retention of meta-knowledge). Experiments demonstrate that SOLAR outperforms strong baselines on common-sense, mathematical, medical, coding, social and logical reasoning tasks, marking a significant step toward autonomous agents capable of lifelong adaptation in evolving environments.", "authors": ["Nitin Vetcha", "Dianbo Liu"], "categories": ["cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2026-03-23", "url": "https://arxiv.org/abs/2605.20189", "pdf_url": "https://arxiv.org/pdf/2605.20189v1", "arxiv_id": "2605.20189", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "CEUR Workshop Proceedings, Vol. 4183, 2026", "quality_score": 0.5179}
{"id": "acd3f2a5770b555b52105cb6a758171a0ead5c50e025295f6fd28fae7f0525cb", "sources": ["arxiv", "semantic_scholar"], "title": "Disentangling Dynamical Systems: Causal Representation Learning Meets Local Sparse Attention", "abstract": "Parametric system identification methods estimate the parameters of explicitly defined physical systems from data. Yet, they remain constrained by the need to provide an explicit function space, typically through a predefined library of candidate functions chosen via available domain knowledge. In contrast, deep learning can demonstrably model systems of broad complexity with high fidelity, but black-box function approximation typically fails to yield explicit descriptive or disentangled representations revealing the structure of a system. We develop a novel identifiability theorem, leveraging causal representation learning, to uncover disentangled representations of system parameters without structural assumptions. We derive a graphical criterion specifying when system parameters can be uniquely disentangled from raw trajectory data, up to permutation and diffeomorphism. Crucially, our analysis demonstrates that global causal structures provide a lower bound on the disentanglement guarantees achievable when considering local state-dependent causal structures. We instantiate system parameter identification as a variational inference problem, leveraging a sparsity-regularised transformer to uncover state-dependent causal structures. We empirically validate our approach across four synthetic domains, demonstrating its ability to recover highly disentangled representations that baselines fail to recover. Corroborating our theoretical analysis, our results confirm that enforcing local causal structure is often necessary for full identifiability.", "authors": ["Markus W. Baumgartner", "Anson Lei", "Joe Watson", "Ingmar Posner"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2026-03-15", "url": "https://arxiv.org/abs/2603.14483", "pdf_url": "https://arxiv.org/pdf/2603.14483v2", "arxiv_id": "2603.14483", "doi": "10.48550/arXiv.2603.14483", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.5088}
{"id": "e4af03ec2aafcfa998e4e76fda481799656b899df77dbce3a736bea26b63cb42", "sources": ["arxiv", "semantic_scholar"], "title": "Chemical Reaction Networks Learn Better than Spiking Neural Networks", "abstract": "We mathematically prove that chemical reaction networks without hidden layers can solve tasks for which spiking neural networks require hidden layers. Our proof uses the deterministic mass-action kinetics formulation of chemical reaction networks. Specifically, we prove that a certain reaction network without hidden layers can learn a classification task previously proved to be achievable by a spiking neural network with hidden layers. We provide analytical regret bounds for the global behavior of the network and analyze its asymptotic behavior and Vapnik-Chervonenkis dimension. In a numerical experiment, we confirm the learning capacity of the proposed chemical reaction network for classifying handwritten digits in pixel images, and we show that it solves the task more accurately and efficiently than a spiking neural network with hidden layers. This provides a motivation for machine learning in chemical computers and a mathematical explanation for how biological cells might exhibit more efficient learning behavior within biochemical reaction networks than neuronal networks.", "authors": ["Sophie Jaffard", "Ivo F. Sbalzarini"], "categories": ["cs.LG", "cs.AI", "math.ST", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2026-03-12", "url": "https://arxiv.org/abs/2603.12060", "pdf_url": "https://arxiv.org/pdf/2603.12060v1", "arxiv_id": "2603.12060", "doi": "10.48550/arXiv.2603.12060", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.5053}
{"id": "5b57e70e78989575921a74f7098879b86f806386aaa83c2d139e79718924d0b9", "sources": ["arxiv", "semantic_scholar"], "title": "Pretrained Vision-Language-Action Models are Surprisingly Resistant to Forgetting in Continual Learning", "abstract": "Continual learning is a long-standing challenge in robot policy learning, where a policy must acquire new skills over time without catastrophically forgetting previously learned ones. While prior work has extensively studied continual learning in relatively small behavior cloning (BC) policy models trained from scratch, its behavior in modern large-scale pretrained Vision-Language-Action (VLA) models remains underexplored. In this work, we found that pretrained VLAs are remarkably resistant to forgetting compared with smaller policy models trained from scratch. Simple Experience Replay (ER) works surprisingly well on VLAs, sometimes achieving zero forgetting even with a small replay data size. Our analysis reveals that pretraining plays a critical role in downstream continual learning performance: large pretrained models mitigate forgetting with a small replay buffer size while maintaining strong forward learning capabilities. Furthermore, we found that VLAs can retain relevant knowledge from prior tasks despite performance degradation during learning new tasks. This knowledge retention enables rapid recovery of seemingly forgotten skills through finetuning. Together, these insights imply that large-scale pretraining fundamentally changes the dynamics of continual learning, enabling models to continually acquire new skills over time with simple replay. Code and more information can be found at https://continual-vlas.github.io/forget-me-not/", "authors": ["Huihan Liu", "Changyeon Kim", "Bo Liu", "Minghuan Liu", "Yuke Zhu"], "categories": ["cs.LG", "cs.AI", "cs.RO"], "fields_of_study": ["Computer Science"], "published_date": "2026-03-04", "url": "https://arxiv.org/abs/2603.03818", "pdf_url": "https://arxiv.org/pdf/2603.03818v2", "arxiv_id": "2603.03818", "doi": "10.48550/arXiv.2603.03818", "citation_count": 4, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4961}
{"id": "b52424d3b29c08887764ee4a08292b7bd5d3e2062869471a17eb19a0adc6d7de", "sources": ["arxiv", "semantic_scholar"], "title": "Why Do Neural Networks Forget: A Study of Collapse in Continual Learning", "abstract": "Catastrophic forgetting is a major problem in continual learning, and lots of approaches arise to reduce it. However, most of them are evaluated through task accuracy, which ignores the internal model structure. Recent research suggests that structural collapse leads to loss of plasticity, as evidenced by changes in effective rank (eRank). This indicates a link to forgetting, since the networks lose the ability to expand their feature space to learn new tasks, which forces the network to overwrite existing representations. Therefore, in this study, we investigate the correlation between forgetting and collapse through the measurement of both weight and activation eRank. To be more specific, we evaluated four architectures, including MLP, ConvGRU, ResNet-18, and Bi-ConvGRU, in the split MNIST and Split CIFAR-100 benchmarks. Those models are trained through the SGD, Learning-without-Forgetting (LwF), and Experience Replay (ER) strategies separately. The results demonstrate that forgetting and collapse are strongly related, and different continual learning strategies help models preserve both capacity and performance in different efficiency.", "authors": ["Yunqin Zhu", "Jun Jin"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2026-03-04", "url": "https://arxiv.org/abs/2603.04580", "pdf_url": "https://arxiv.org/pdf/2603.04580v1", "arxiv_id": "2603.04580", "doi": "10.48550/arXiv.2603.04580", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4961}
{"id": "9d4dd059a647f0af15a08ed4eceebaed33f751c61372a6522ca43e5c0a77ecc2", "sources": ["arxiv", "semantic_scholar"], "title": "Position: Modular Memory is the Key to Continual Learning Agents", "abstract": "Foundation models have transformed machine learning through large-scale pretraining and increased test-time compute. Despite surpassing human performance in several domains, these models remain fundamentally limited in continuous operation, experience accumulation, and personalization, capabilities that are central to adaptive intelligence. While continual learning research has long targeted these goals, its historical focus on in-weight learning (IWL), i.e., updating a single model's parameters to absorb new knowledge, has rendered catastrophic forgetting a persistent challenge. Our position is that combining the strengths of In-Weight Learning (IWL) and the newly emerged capabilities of In-Context Learning (ICL) through the design of modular memory is the missing piece for continual adaptation at scale. We outline a conceptual framework for modular memory-centric architectures that leverage ICL for rapid adaptation and knowledge accumulation, and IWL for stable updates to model capabilities, charting a practical roadmap toward continually learning agents.", "authors": ["Vaggelis Dorovatas", "Malte Schwerin", "Andrew D. Bagdanov", "Lucas Caccia", "Antonio Carta", "Laurent Charlin", "Barbara Hammer", "Tyler L. Hayes", "Timm Hess", "Christopher Kanan", "Dhireesha Kudithipudi", "Xialei Liu", "Vincenzo Lomonaco", "Jorge Mendez-Mendez", "Darshan Patil", "Ameya Prabhu", "Elisa Ricci", "Tinne Tuytelaars", "Gido M. van de Ven", "Liyuan Wang", "Joost van de Weijer", "Jonghyun Choi", "Martin Mundt", "Rahaf Aljundi"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2026-03-02", "url": "https://arxiv.org/abs/2603.01761", "pdf_url": "https://arxiv.org/pdf/2603.01761v2", "arxiv_id": "2603.01761", "doi": "10.48550/arXiv.2603.01761", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4939}
{"id": "1e1560bb31fac79cd470c3115d6885bf4ce92f1e1b2d538f66384209632be9bd", "sources": ["arxiv", "semantic_scholar"], "title": "Learning in the Null Space: Small Singular Values for Continual Learning", "abstract": "Alleviating catastrophic forgetting while enabling further learning is a primary challenge in continual learning (CL). Orthogonal-based training methods have gained attention for their efficiency and strong theoretical properties, and many existing approaches enforce orthogonality through gradient projection. In this paper, we revisit orthogonality and exploit the fact that small singular values correspond to directions that are nearly orthogonal to the input space of previous tasks. Building on this principle, we introduce NESS (Null-space Estimated from Small Singular values), a CL method that applies orthogonality directly in the weight space rather than through gradient manipulation. Specifically, NESS constructs an approximate null space using the smallest singular values of each layer's input representation and parameterizes task-specific updates via a compact low-rank adaptation (LoRA-style) formulation constrained to this subspace. The subspace basis is fixed to preserve the null-space constraint, and only a single trainable matrix is learned for each task. This design ensures that the resulting updates remain approximately in the null space of previous inputs while enabling adaptation to new tasks. Our theoretical analysis and experiments on three benchmark datasets demonstrate competitive performance, low forgetting, and stable accuracy across tasks, highlighting the role of small singular values in continual learning. The code is available at https://github.com/pacman-ctm/NESS.", "authors": ["Cuong Anh Pham", "Praneeth Vepakomma", "Samuel Horváth"], "categories": ["cs.LG", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2026-02-25", "url": "https://arxiv.org/abs/2602.21919", "pdf_url": "https://arxiv.org/pdf/2602.21919v1", "arxiv_id": "2602.21919", "doi": "10.48550/arXiv.2602.21919", "citation_count": 0, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/pacman-ctm/NESS", "venue": "arXiv.org", "quality_score": 0.7544}
{"id": "17d5f96ef84969f5a386fee224cdef9af4ecb2612d7d9e0a9c31a0fc1dd987b1", "sources": ["arxiv", "semantic_scholar"], "title": "Value Bonuses using Ensemble Errors for Exploration in Reinforcement Learning", "abstract": "Optimistic value estimates provide one mechanism for directed exploration in reinforcement learning (RL). The agent acts greedily with respect to an estimate of the value plus what can be seen as a value bonus. The value bonus can be learned by estimating a value function on reward bonuses, propagating local uncertainties around rewards. However, this approach only increases the value bonus for an action retroactively, after seeing a higher reward bonus from that state and action. Such an approach does not encourage the agent to visit a state and action for the first time. In this work, we introduce an algorithm for exploration called Value Bonuses with Ensemble errors (VBE), that maintains an ensemble of random action-value functions (RQFs). VBE uses the errors in the estimation of these RQFs to design value bonuses that provide first-visit optimism and deep exploration. The key idea is to design the rewards for these RQFs in such a way that the value bonus can decrease to zero. We show that VBE outperforms Bootstrap DQN and two reward bonus approaches (RND and ACB) on several classic environments used to test exploration and provide demonstrative experiments that it can scale easily to more complex environments like Atari.", "authors": ["Abdul Wahab", "Raksha Kumaraswamy", "Martha White"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2026-02-12", "url": "https://arxiv.org/abs/2602.12375", "pdf_url": "https://arxiv.org/pdf/2602.12375v1", "arxiv_id": "2602.12375", "doi": "10.48550/arXiv.2602.12375", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4732}
{"id": "d790c88703165f0f1bb398150b9539961ceca2247724ee68a9c4f55d5239f3bd", "sources": ["arxiv", "semantic_scholar"], "title": "MerLin: A Discovery Engine for Photonic and Hybrid Quantum Machine Learning", "abstract": "Identifying where quantum models may offer practical benefits in near term quantum machine learning (QML) requires moving beyond isolated algorithmic proposals toward systematic and empirical exploration across models, datasets, and hardware constraints. We introduce MerLin, an open-source framework designed as a discovery engine for photonic and hybrid quantum machine learning. MerLin integrates optimized strong simulation of linear optical circuits into standard PyTorch and scikit learn workflows, enabling end-to-end differentiable training of quantum layers. MerLin is designed around systematic benchmarking and reproducibility. As an initial contribution, we reproduce eighteen state-of-the-art photonic and hybrid QML works spanning kernel methods, reservoir computing, convolutional and recurrent architectures, generative models, and modern training paradigms. These reproductions are released as reusable, modular experiments that can be directly extended and adapted, establishing a shared experimental baseline consistent with empirical benchmarking methodologies widely adopted in modern artificial intelligence. By embedding photonic quantum models within established machine learning ecosystems, MerLin allows practitioners to leverage existing tooling for ablation studies, cross-modality comparisons, and hybrid classical-quantum workflows. The framework already implements hardware-aware features, allowing tests on available quantum hardware while enabling exploration beyond its current capabilities, positioning MerLin as a forward-looking co-design tool linking algorithms, benchmarks, and hardware.", "authors": ["Cassandre Notton", "Benjamin Stott", "Philippe Schoeb", "Anthony Walsh", "Grégoire Leboucher", "Vincent Espitalier", "Vassilis Apostolou", "Louis-Félix Vigneux", "Alexia Salavrakos", "Jean Senellart"], "categories": ["cs.LG", "cs.PL", "quant-ph"], "fields_of_study": ["Computer Science", "Physics"], "published_date": "2026-02-11", "url": "https://arxiv.org/abs/2602.11092", "pdf_url": "https://arxiv.org/pdf/2602.11092v2", "arxiv_id": "2602.11092", "doi": "10.48550/arXiv.2602.11092", "citation_count": 1, "influential_citation_count": 0, "has_code": true, "code_url": null, "venue": "arXiv.org", "quality_score": 0.7296}
{"id": "0d014a287f0d92d1d96ea9be47e66efbad5a22a27255bed97f6e39c24a437ac4", "sources": ["arxiv", "semantic_scholar"], "title": "A Thermodynamic Theory of Learning Part II: Critical Period Closure and Continual Learning Failure", "abstract": "Learning performed over finite time is inherently irreversible. In Part~I of this series, we modeled learning as a transport process in the space of parameter distributions and derived the Epistemic Speed Limit (ESL), which lower-bounds entropy production under finite-time dynamics. In this work (Part~II), we show that irreversibility imposes a geometric restriction on future adaptability through the compositional structure of learning dynamics. Successive learning phases compose multiplicatively as transport maps, and their Jacobians form a semigroup whose rank and singular values are submultiplicative. As a result, dynamically usable degrees of reconfiguration can only decrease under composition. We formalize the remaining adaptability of a model in terms of compatible effective rank, defined as the log-volume of task-preserving directions that remain dynamically accessible. Although task performance may remain unchanged, finite-time learning can progressively reduce this reconfiguration capacity. We prove a capacity-threshold criterion for continual learning: let m_B denote the stable rank of the Hessian of a new task B restricted to the task-preserving manifold of a previously learned task A. If m_B exceeds the residual compatible effective rank, then task B is trajectory-level incompatible with task A; any sufficient adaptation necessarily induces forgetting. Thus catastrophic forgetting arises not from the absence of multi-task solutions, but from irreversible loss of reconfiguration capacity under compositional learning dynamics. This establishes a trajectory-level capacity limit for continual learning.", "authors": ["Daisuke Okanohara"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2026-02-08", "url": "https://arxiv.org/abs/2602.07950", "pdf_url": "https://arxiv.org/pdf/2602.07950v2", "arxiv_id": "2602.07950", "doi": "10.48550/arXiv.2602.07950", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4686}
{"id": "8e0f0b27f0ed9c58c12203cab5c956ac0cc9f12460329a9b1ff2499be9353e3a", "sources": ["arxiv", "semantic_scholar"], "title": "Keep Rehearsing and Refining: Lifelong Learning Vehicle Routing under Continually Drifting Tasks", "abstract": "Existing neural solvers for vehicle routing problems (VRPs) are typically trained either in a one-off manner on a fixed set of pre-defined tasks or in a lifelong manner with tasks arriving sequentially, assuming sufficient training on each task. Both settings overlook a common real-world property: problem patterns may drift continually over time, yielding massive tasks sequentially arising, each with only limited training resources. In this paper, we propose a novel lifelong learning paradigm for neural VRP solvers under continual task drift over time, where each task is locally stationary at one learning time step but receives only insufficient training resources. We empirically demonstrate that such continual drift arises in practice using a real-world logistics dataset. We then propose Dual Replay with Experience Enhancement (DREE), a general framework to improve learning efficiency and mitigate catastrophic forgetting under such drift. Extensive experiments based on both the real-world logistics dataset and commonly used synthetic dataset show that, under such continual drift, DREE effectively learns new tasks, preserves prior knowledge, improves generalization to unseen tasks, and can be applied to various existing neural solvers.", "authors": ["Jiyuan Pei", "Yi Mei", "Jialin Liu", "Mengjie Zhang", "Xin Yao"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2026-01-30", "url": "https://arxiv.org/abs/2601.22509", "pdf_url": "https://arxiv.org/pdf/2601.22509v2", "arxiv_id": "2601.22509", "doi": "10.48550/arXiv.2601.22509", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4583}
{"id": "7e6a12b9aa29842f32aef94bba1255f3e4f70ecfa69a5647e64d254b6f339d22", "sources": ["arxiv", "semantic_scholar"], "title": "Federated Learning Under Temporal Drift -- Mitigating Catastrophic Forgetting via Experience Replay", "abstract": "Federated Learning struggles under temporal concept drift where client data distributions shift over time. We demonstrate that standard FedAvg suffers catastrophic forgetting under seasonal drift on Fashion-MNIST, with accuracy dropping from 74% to 28%. We propose client-side experience replay, where each client maintains a small buffer of past samples mixed with current data during local training. This simple approach requires no changes to server aggregation. Experiments show that a 50-sample-per-class buffer restores performance to 78-82%, effectively preventing forgetting. Our ablation study reveals a clear memory-accuracy trade-off as buffer size increases.", "authors": ["Sahasra Kokkula", "Daniel David", "Aaditya Baruah"], "categories": ["cs.LG", "cs.DC"], "fields_of_study": ["Computer Science"], "published_date": "2026-01-19", "url": "https://arxiv.org/abs/2601.13456", "pdf_url": "https://arxiv.org/pdf/2601.13456v1", "arxiv_id": "2601.13456", "doi": "10.48550/arXiv.2601.13456", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4457}
{"id": "5914cf9b44fd29434dcce3c96c191fc84d4853b285a8bb4fafaced78ba3750d8", "sources": ["arxiv", "semantic_scholar"], "title": "Exploring Student Expectations and Confidence in Learning Analytics", "abstract": "Learning Analytics (LA) is nowadays ubiquitous in many educational systems, providing the ability to collect and analyze student data in order to understand and optimize learning and the environments in which it occurs. On the other hand, the collection of data requires to comply with the growing demand regarding privacy legislation. In this paper, we use the Student Expectation of Learning Analytics Questionnaire (SELAQ) to analyze the expectations and confidence of students from different faculties regarding the processing of their data for Learning Analytics purposes. This allows us to identify four clusters of students through clustering algorithms: Enthusiasts, Realists, Cautious and Indifferents. This structured analysis provides valuable insights into the acceptance and criticism of Learning Analytics among students.", "authors": ["Hayk Asatryan", "Basile Tousside", "Janis Mohr", "Malte Neugebauer", "Hildo Bijl", "Paul Spiegelberg", "Claudia Frohn-Schauf", "Jörg Frochte"], "categories": ["cs.LG", "cs.CY", "cs.HC"], "fields_of_study": ["Computer Science"], "published_date": "2026-01-08", "url": "https://arxiv.org/abs/2601.05082", "pdf_url": "https://arxiv.org/pdf/2601.05082v1", "arxiv_id": "2601.05082", "doi": "10.1145/3636555.3636923", "citation_count": 2, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "International Conference on Learning Analytics and Knowledge", "quality_score": 0.4331}
{"id": "618492aa2a53a0f44392593bafa31587bdcec2b909b87ab09332a63b00b16651", "sources": ["arxiv", "semantic_scholar"], "title": "Shallow Neural Networks Learn Low-Degree Spherical Polynomials with Feature Learning by Learnable Channel Attention", "abstract": "We study the problem of learning a low-degree spherical polynomial of degree $\\ell_0 = Θ(1) \\ge 1$ defined on the unit sphere in $\\RR^d$ by training an over-parameterized two-layer neural network (NN) with channel attention in this paper. Our main result is the significantly improved sample complexity for learning such low-degree polynomials. We show that, for any regression risk $\\eps \\in (0,1)$, a carefully designed two-layer NN with channel attention and finite width trained by the vanilla gradient descent (GD) requires the lowest sample complexity of $n \\asymp Θ(d^{\\ell_0}/\\eps)$ with high probability, in contrast with the representative sample complexity $Θ\\pth{d^{\\ell_0} \\max\\set{\\eps^{-2},\\log d}}$, where $n$ is the training data size. Moreover, such sample complexity is not improvable since the trained network renders a sharp rate of the nonparametric regression risk of the order $Θ(d^{\\ell_0}/{n})$ with high probability. On the other hand, the minimax optimal rate for the regression risk with a kernel of rank $Θ(d^{\\ell_0})$ is $Θ(d^{\\ell_0}/{n})$, so that the rate of the nonparametric regression risk of the network trained by GD is minimax optimal. Training the two-layer NN with channel attention proceeds in two stages: (1) a provable learnable channel selection algorithm, as a learnable harmonic-degree selection process, identifies the ground truth channel number in the target function, $\\ell_0$, from $L \\ge \\ell_0$ channels in the first-layer activation; (2) the second layer is trained by standard GD using the selected channels. To the best of our knowledge, this is the first time a minimax optimal risk bound is obtained by training an over-parameterized but finite-width neural network with feature learning capability to learn low-degree spherical polynomials.", "authors": ["Yingzhen Yang"], "categories": ["stat.ML", "cs.LG", "math.OC"], "fields_of_study": ["Mathematics", "Computer Science"], "published_date": "2025-12-23", "url": "https://arxiv.org/abs/2512.20562", "pdf_url": "https://arxiv.org/pdf/2512.20562v2", "arxiv_id": "2512.20562", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.264}
{"id": "74eaccfd7edd9bf62a60b0f99b0db8b1cd74e122db4d24887b93c9a6a9ede1d7", "sources": ["arxiv", "semantic_scholar"], "title": "Sequencing to Mitigate Catastrophic Forgetting in Continual Learning", "abstract": "To cope with real-world dynamics, an intelligent system needs to incrementally acquire, update, and exploit knowledge throughout its lifetime. This ability, known as Continual learning, provides a foundation for AI systems to develop themselves adaptively. Catastrophic forgetting is a major challenge to the progress of Continual Learning approaches, where learning a new task usually results in a dramatic performance drop on previously learned ones. Many approaches have emerged to counteract the impact of CF. Most of the proposed approaches can be categorized into five classes: replay-based, regularization-based, optimization-based, representation-based, and architecture-based. In this work, we approach the problem from a different angle, specifically by considering the optimal sequencing of tasks as they are presented to the model. We investigate the role of task sequencing in mitigating CF and propose a method for determining the optimal task order. The proposed method leverages zero-shot scoring algorithms inspired by neural architecture search (NAS). Results demonstrate that intelligent task sequencing can substantially reduce CF. Moreover, when combined with traditional continual learning strategies, sequencing offers enhanced performance and robustness against forgetting. Additionally, the presented approaches can find applications in other fields, such as curriculum learning.", "authors": ["Hesham G. Moussa", "Aroosa Hameed", "Arashmid Akhavain"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-12-18", "url": "https://arxiv.org/abs/2512.16871", "pdf_url": "https://arxiv.org/pdf/2512.16871v1", "arxiv_id": "2512.16871", "doi": "10.48550/arXiv.2512.16871", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4091}
{"id": "78d4cae4f641146f4bbd6c08318b815ca2bd8780258e1a2a811ff630a80cd334", "sources": ["arxiv", "semantic_scholar"], "title": "Continual Learning at the Edge: An Agnostic IIoT Architecture", "abstract": "The exponential growth of Internet-connected devices has presented challenges to traditional centralized computing systems due to latency and bandwidth limitations. Edge computing has evolved to address these difficulties by bringing computations closer to the data source. Additionally, traditional machine learning algorithms are not suitable for edge-computing systems, where data usually arrives in a dynamic and continual way. However, incremental learning offers a good solution for these settings. We introduce a new approach that applies the incremental learning philosophy within an edge-computing scenario for the industrial sector with a specific purpose: real time quality control in a manufacturing system. Applying continual learning we reduce the impact of catastrophic forgetting and provide an efficient and effective solution.", "authors": ["Pablo García-Santaclara", "Bruno Fernández-Castro", "Rebeca P. Díaz-Redondo", "Carlos Calvo-Moa", "Henar Mariño-Bodelón"], "categories": ["stat.ML", "cs.LG"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2025-12-16", "url": "https://arxiv.org/abs/2512.14311", "pdf_url": "https://arxiv.org/pdf/2512.14311v1", "arxiv_id": "2512.14311", "doi": "10.1007/978-981-96-6938-7_33", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4068}
{"id": "14b4c3e551bff6b7a8eca4df632d968f6504f9170ef77f3b0f3c36ff7e3ad8d9", "sources": ["arxiv", "semantic_scholar"], "title": "Multiclass Graph-Based Large Margin Classifiers: Unified Approach for Support Vectors and Neural Networks", "abstract": "While large margin classifiers are originally an outcome of an optimization framework, support vectors (SVs) can be obtained from geometric approaches. This article presents advances in the use of Gabriel graphs (GGs) in binary and multiclass classification problems. For Chipclass, a hyperparameter-less and optimization-less GG-based binary classifier, we discuss how activation functions and support edge (SE)-centered neurons affect the classification, proposing smoother functions and structural SV (SSV)-centered neurons to achieve margins with low probabilities and smoother classification contours. We extend the neural network architecture, which can be trained with backpropagation with a softmax function and a cross-entropy loss, or by solving a system of linear equations. A new subgraph-/distance-based membership function for graph regularization is also proposed, along with a new GG recomputation algorithm that is less computationally expensive than the standard approach. Experimental results with the Friedman test show that our method was better than previous GG-based classifiers and statistically equivalent to tree-based models.", "authors": ["Vítor M. Hanriot", "Luiz C. B. Torres", "Antônio P. Braga"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Medicine", "Mathematics"], "published_date": "2025-12-15", "url": "https://arxiv.org/abs/2512.13410", "pdf_url": "https://arxiv.org/pdf/2512.13410v1", "arxiv_id": "2512.13410", "doi": "10.1109/TNNLS.2024.3420227", "citation_count": 3, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.4056}
{"id": "455a5c2cb4fe3c769f06beaf1267f20415aff21f777759daf1a0dbfb87c43a0f", "sources": ["arxiv", "semantic_scholar"], "title": "Bridging Streaming Continual Learning via In-Context Large Tabular Models", "abstract": "In streaming scenarios, models must learn continuously, adapting to concept drifts without erasing previously acquired knowledge. However, existing research communities address these challenges in isolation. Continual Learning (CL) focuses on long-term retention and mitigating catastrophic forgetting, often without strict real-time constraints. Stream Learning (SL) emphasizes rapid, efficient adaptation to high-frequency data streams, but typically neglects forgetting. Recent efforts have tried to combine these paradigms, yet no clear algorithmic overlap exists. We argue that large in-context tabular models (LTMs) provide a natural bridge for Streaming Continual Learning (SCL). In our view, unbounded streams should be summarized on-the-fly into compact sketches that can be consumed by LTMs. This recovers the classical SL motivation of compressing massive streams with fixed-size guarantees, while simultaneously aligning with the experience-replay desiderata of CL. To clarify this bridge, we show how the SL and CL communities implicitly adopt a divide-to-conquer strategy to manage the tension between plasticity (performing well on the current distribution) and stability (retaining past knowledge), while also imposing a minimal complexity constraint that motivates diversification (avoiding redundancy in what is stored) and retrieval (re-prioritizing past information when needed). Within this perspective, we propose structuring SCL with LTMs around two core principles of data selection for in-context learning: (1) distribution matching, which balances plasticity and stability, and (2) distribution compression, which controls memory size through diversification and retrieval mechanisms.", "authors": ["Afonso Lourenço", "João Gama", "Eric P. Xing", "Goreti Marreiros"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-12-12", "url": "https://arxiv.org/abs/2512.11668", "pdf_url": "https://arxiv.org/pdf/2512.11668v1", "arxiv_id": "2512.11668", "doi": "10.48550/arXiv.2512.11668", "citation_count": 4, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2559}
{"id": "4bb3a6793a83c3c7835a15ba82993513caa574d4efea3991530e94b926a5a78d", "sources": ["arxiv", "semantic_scholar"], "title": "Angular Regularization for Positive-Unlabeled Learning on the Hypersphere", "abstract": "Positive-Unlabeled (PU) learning addresses classification problems where only a subset of positive examples is labeled and the remaining data is unlabeled, making explicit negative supervision unavailable. Existing PU methods often rely on negative-risk estimation or pseudo-labeling, which either require strong distributional assumptions or can collapse in high-dimensional settings. We propose AngularPU, a novel PU framework that operates on the unit hypersphere using cosine similarity and angular margin. In our formulation, the positive class is represented by a learnable prototype vector, and classification reduces to thresholding the cosine similarity between an embedding and this prototype-eliminating the need for explicit negative modeling. To counteract the tendency of unlabeled embeddings to cluster near the positive prototype, we introduce an angular regularizer that encourages dispersion of the unlabeled set over the hypersphere, improving separation. We provide theoretical guarantees on the Bayes-optimality of the angular decision rule, consistency of the learned prototype, and the effect of the regularizer on the unlabeled distribution. Experiments on benchmark datasets demonstrate that AngularPU achieves competitive or superior performance compared to state-of-the-art PU methods, particularly in settings with scarce positives and high-dimensional embeddings, while offering geometric interpretability and scalability.", "authors": ["Vasileios Sevetlidis", "George Pavlidis", "Antonios Gasteratos"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-12-07", "url": "https://arxiv.org/abs/2512.06785", "pdf_url": "https://arxiv.org/pdf/2512.06785v1", "arxiv_id": "2512.06785", "doi": "10.48550/arXiv.2512.06785", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Transactions on Machine Learning Research, 2025", "quality_score": 0.3965}
{"id": "09e986ece22d720241038e4e1760c1feab3bd4a18b9d0b2e58f3d5d6c84ea75d", "sources": ["arxiv", "semantic_scholar"], "title": "Mitigating Catastrophic Forgetting in Mathematical Reasoning Finetuning through Mixed Training", "abstract": "When finetuning large language models for specialized tasks such as mathematical reasoning, models exhibit catastrophic forgetting, losing previously learned capabilities. We investigate this by finetuning Flan-T5-Base (250M parameters) on the DeepMind Mathematics dataset and measuring forgetting on MultiNLI. Math-only training improves mathematical accuracy from 3.1\\% to 12.0\\% but causes NLI accuracy to collapse from 81.0\\% to 16.5\\%--a 64.5 percentage point drop occurring within the first 1,000 training steps. We propose mixed training strategies that interleave mathematical and NLI examples during training. Our results demonstrate that mixed training completely eliminates catastrophic forgetting while maintaining equivalent mathematical performance: the balanced 1:1 ratio achieves 12.0\\% math accuracy (matching math-only) while preserving 86.2\\% NLI accuracy. We systematically explore mixing ratios from 1:1 to 15:1, finding that even minimal NLI exposure (6.2\\%) provides effective regularization. These findings demonstrate that specialization need not require forgetting general capabilities, with implications for scaling to larger models where mixed training may confer additional benefits beyond forgetting prevention.", "authors": ["John Graham Reynolds"], "categories": ["cs.LG", "cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-12-05", "url": "https://arxiv.org/abs/2512.13706", "pdf_url": "https://arxiv.org/pdf/2512.13706v1", "arxiv_id": "2512.13706", "doi": "10.48550/arXiv.2512.13706", "citation_count": 0, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/johngrahamreynolds/mathematical_catastrophe_mitigation", "venue": "arXiv.org", "quality_score": 0.6092}
{"id": "73f4e5ea821d529d5ee0b07719a06076fe00b43a43629f1eb7599a437d2766f6", "sources": ["arxiv", "semantic_scholar"], "title": "Sample Complexity of Distributionally Robust Off-Dynamics Reinforcement Learning with Online Interaction", "abstract": "Off-dynamics reinforcement learning (RL), where training and deployment transition dynamics are different, can be formulated as learning in a robust Markov decision process (RMDP) where uncertainties in transition dynamics are imposed. Existing literature mostly assumes access to generative models allowing arbitrary state-action queries or pre-collected datasets with a good state coverage of the deployment environment, bypassing the challenge of exploration. In this work, we study a more realistic and challenging setting where the agent is limited to online interaction with the training environment. To capture the intrinsic difficulty of exploration in online RMDPs, we introduce the supremal visitation ratio, a novel quantity that measures the mismatch between the training dynamics and the deployment dynamics. We show that if this ratio is unbounded, online learning becomes exponentially hard. We propose the first computationally efficient algorithm that achieves sublinear regret in online RMDPs with $f$-divergence based transition uncertainties. We also establish matching regret lower bounds, demonstrating that our algorithm achieves optimal dependence on both the supremal visitation ratio and the number of interaction episodes. Finally, we validate our theoretical results through comprehensive numerical experiments.", "authors": ["Yiting He", "Zhishuai Liu", "Weixin Wang", "Pan Xu"], "categories": ["cs.LG", "cs.AI", "cs.RO", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2025-11-07", "url": "https://arxiv.org/abs/2511.05396", "pdf_url": "https://arxiv.org/pdf/2511.05396v1", "arxiv_id": "2511.05396", "doi": "10.48550/arXiv.2511.05396", "citation_count": 13, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.3621}
{"id": "4f6770d70b203cd13be6e63f9ff466ab2a3e1de424619d79997453fe6709ce19", "sources": ["arxiv", "semantic_scholar"], "title": "Forgetting is Everywhere", "abstract": "A fundamental challenge in developing general learning algorithms is their tendency to forget past knowledge when adapting to new data. Addressing this problem requires a principled understanding of forgetting; yet, despite decades of study, no unified definition has emerged that provides insights into the underlying dynamics of learning. We propose an algorithm- and task-agnostic theory that characterises forgetting as a lack of self-consistency in a learner's predictive distribution, manifesting as a loss of predictive information. Our theory naturally yields a general measure of an algorithm's propensity to forget and demonstrates that exact Bayesian inference allows for adaptation without forgetting. To validate the theory, we design a comprehensive set of experiments that span classification, regression, generative modelling, and reinforcement learning. We empirically demonstrate how forgetting is present across all deep learning settings and plays a significant role in determining learning efficiency. Together, these results establish a principled understanding of forgetting and lay the foundation for analysing and improving the information retention capabilities of general learning algorithms.", "authors": ["Ben Sanati", "Thomas L. Lee", "Trevor McInroe", "Aidan Scannell", "Nikolay Malkin", "David Abel", "Amos Storkey"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2025-11-06", "url": "https://arxiv.org/abs/2511.04666", "pdf_url": "https://arxiv.org/pdf/2511.04666v3", "arxiv_id": "2511.04666", "doi": "10.48550/arXiv.2511.04666", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3609}
{"id": "78465d6e36b164bbf07bf6119676da7d0c8dee742fe618e5cb8208a674b31f5d", "sources": ["arxiv", "semantic_scholar"], "title": "Benchmarking Catastrophic Forgetting Mitigation Methods in Federated Time Series Forecasting", "abstract": "Catastrophic forgetting (CF) poses a persistent challenge in continual learning (CL), especially within federated learning (FL) environments characterized by non-i.i.d. time series data. While existing research has largely focused on classification tasks in vision domains, the regression-based forecasting setting prevalent in IoT and edge applications remains underexplored. In this paper, we present the first benchmarking framework tailored to investigate CF in federated continual time series forecasting. Using the Beijing Multi-site Air Quality dataset across 12 decentralized clients, we systematically evaluate several CF mitigation strategies, including Replay, Elastic Weight Consolidation, Learning without Forgetting, and Synaptic Intelligence. Key contributions include: (i) introducing a new benchmark for CF in time series FL, (ii) conducting a comprehensive comparative analysis of state-of-the-art methods, and (iii) releasing a reproducible open-source framework. This work provides essential tools and insights for advancing continual learning in federated time-series forecasting systems.", "authors": ["Khaled Hallak", "Oudom Kem"], "categories": ["cs.LG", "cs.DC", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2025-10-24", "url": "https://arxiv.org/abs/2510.21491", "pdf_url": "https://arxiv.org/pdf/2510.21491v1", "arxiv_id": "2510.21491", "doi": "10.1109/FLTA67013.2025.11336525", "citation_count": 0, "influential_citation_count": 0, "has_code": true, "code_url": null, "venue": null, "quality_score": 0.409}
{"id": "24dce3c2adbb9c65aaa655e01bffd3d5f34dc85f2ff8b833915c0620a1da29c2", "sources": ["arxiv", "semantic_scholar"], "title": "Multi-modal Co-learning for Earth Observation: Enhancing single-modality models via modality collaboration", "abstract": "Multi-modal co-learning is emerging as an effective paradigm in machine learning, enabling models to collaboratively learn from different modalities to enhance single-modality predictions. Earth Observation (EO) represents a quintessential domain for multi-modal data analysis, wherein diverse remote sensors collect data to sense our planet. This unprecedented volume of data introduces novel challenges. Specifically, the access to the same sensor modalities at both training and inference stages becomes increasingly complex based on real-world constraints affecting remote sensing platforms. In this context, multi-modal co-learning presents a promising strategy to leverage the vast amount of sensor-derived data available at the training stage to improve single-modality models for inference-time deployment. Most current research efforts focus on designing customized solutions for either particular downstream tasks or specific modalities available at the inference stage. To address this, we propose a novel multi-modal co-learning framework capable of generalizing across various tasks without targeting a specific modality for inference. Our approach combines contrastive and modality discriminative learning together to guide single-modality models to structure the internal model manifold into modality-shared and modality-specific information. We evaluate our framework on four EO benchmarks spanning classification and regression tasks across different sensor modalities, where only one of the modalities available during training is accessible at inference time. Our results demonstrate consistent predictive improvements over state-of-the-art approaches from the recent machine learning and computer vision literature, as well as EO-specific methods. The obtained findings validate our framework in the single-modality inference scenarios across a diverse range of EO applications.", "authors": ["Francisco Mena", "Dino Ienco", "Cassio F. Dantas", "Roberto Interdonato", "Andreas Dengel"], "categories": ["cs.CV", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-10-22", "url": "https://arxiv.org/abs/2510.19579", "pdf_url": "https://arxiv.org/pdf/2510.19579v1", "arxiv_id": "2510.19579", "doi": "10.1007/s10994-025-06903-0", "citation_count": 3, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Machine-mediated learning", "quality_score": 0.3438}
{"id": "c9c31f691994cefa3ff3ab942e99940a4c7d6887f166245009e86162c9616008", "sources": ["arxiv", "semantic_scholar"], "title": "On the Implicit Adversariality of Catastrophic Forgetting in Deep Continual Learning", "abstract": "Continual learning seeks the human-like ability to accumulate new skills in machine intelligence. Its central challenge is catastrophic forgetting, whose underlying cause has not been fully understood for deep networks. In this paper, we demystify catastrophic forgetting by revealing that the new-task training is implicitly an adversarial attack against the old-task knowledge. Specifically, the new-task gradients automatically and accurately align with the sharp directions of the old-task loss landscape, rapidly increasing the old-task loss. This adversarial alignment is intriguingly counter-intuitive because the sharp directions are too sparsely distributed to align with by chance. To understand it, we theoretically show that it arises from training's low-rank bias, which, through forward and backward propagation, confines the two directions into the same low-dimensional subspace, facilitating alignment. Gradient projection (GP) methods, a representative family of forgetting-mitigating methods, reduce adversarial alignment caused by forward propagation, but cannot address the alignment due to backward propagation. We propose backGP to address it, which reduces forgetting by 10.8% and improves accuracy by 12.7% on average over GP methods.", "authors": ["Ze Peng", "Jian Zhang", "Jintao Guo", "Lei Qi", "Yang Gao", "Yinghuan Shi"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-10-10", "url": "https://arxiv.org/abs/2510.09181", "pdf_url": "https://arxiv.org/pdf/2510.09181v1", "arxiv_id": "2510.09181", "doi": "10.48550/arXiv.2510.09181", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.33}
{"id": "663a026587d5a7fc9cffbbe310958f18dd9c73027725b813c7c3de8b4af82c1f", "sources": ["arxiv", "semantic_scholar"], "title": "On the Theory of Continual Learning with Gradient Descent for Neural Networks", "abstract": "Continual learning, the ability of a model to adapt to an ongoing sequence of tasks without forgetting earlier ones, is a central goal of artificial intelligence. To better understand its underlying mechanisms, we study the limitations of continual learning in a tractable yet representative setting. Specifically, we analyze one-hidden-layer quadratic neural networks trained by gradient descent on a sequence of XOR-cluster datasets with Gaussian noise, where different tasks correspond to clusters with orthogonal means. Our analysis is based on a tight characterization of gradient descent dynamics for the training loss, which yields explicit bounds on the rate of train-time forgetting as functions of the number of iterations, sample size, number of tasks, and hidden-layer width. We then leverage an algorithmic stability framework to bound the generalization gap, leading to corresponding guarantees on test-time forgetting. Together, our results provide the first closed-form guarantees for forgetting in continual learning with neural networks and show how key problem parameters jointly govern forgetting dynamics. Numerical experiments corroborate our theoretical results.", "authors": ["Hossein Taheri", "Avishek Ghosh", "Arya Mazumdar"], "categories": ["stat.ML", "cs.IT", "cs.LG"], "fields_of_study": ["Mathematics", "Computer Science"], "published_date": "2025-10-07", "url": "https://arxiv.org/abs/2510.05573", "pdf_url": "https://arxiv.org/pdf/2510.05573v2", "arxiv_id": "2510.05573", "doi": "10.48550/arXiv.2510.05573", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3266}
{"id": "97b8b6b1386b59c7ee8404c91f2ce44e69791866e8d2225c6bb562d2599b411b", "sources": ["arxiv", "semantic_scholar"], "title": "Learning Time-Series Representations by Hierarchical Uniformity-Tolerance Latent Balancing", "abstract": "We propose TimeHUT, a novel method for learning time-series representations by hierarchical uniformity-tolerance balancing of contrastive representations. Our method uses two distinct losses to learn strong representations with the aim of striking an effective balance between uniformity and tolerance in the embedding space. First, TimeHUT uses a hierarchical setup to learn both instance-wise and temporal information from input time-series. Next, we integrate a temperature scheduler within the vanilla contrastive loss to balance the uniformity and tolerance characteristics of the embeddings. Additionally, a hierarchical angular margin loss enforces instance-wise and temporal contrast losses, creating geometric margins between positive and negative pairs of temporal sequences. This approach improves the coherence of positive pairs and their separation from the negatives, enhancing the capture of temporal dependencies within a time-series sample. We evaluate our approach on a wide range of tasks, namely 128 UCR and 30 UAE datasets for univariate and multivariate classification, as well as Yahoo and KPI datasets for anomaly detection. The results demonstrate that TimeHUT outperforms prior methods by considerable margins on classification, while obtaining competitive results for anomaly detection. Finally, detailed sensitivity and ablation studies are performed to evaluate different components and hyperparameters of our method.", "authors": ["Amin Jalali", "Milad Soltany", "Michael Greenspan", "Ali Etemad"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-10-02", "url": "https://arxiv.org/abs/2510.01658", "pdf_url": "https://arxiv.org/pdf/2510.01658v1", "arxiv_id": "2510.01658", "doi": "10.48550/arXiv.2510.01658", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Transactions on Machine Learning Research (10/2025)", "quality_score": 0.3208}
{"id": "3ead865678e382e9f9745cbc75e915bc4cf93e0e51bac3f20204cdb5e4dfaad8", "sources": ["arxiv", "semantic_scholar"], "title": "Generative Evolutionary Meta-Solver (GEMS): Scalable Surrogate-Free Multi-Agent Reinforcement Learning", "abstract": "Scalable multi-agent reinforcement learning (MARL) remains a central challenge for AI. Existing population-based methods, like Policy-Space Response Oracles, PSRO, require storing explicit policy populations and constructing full payoff matrices, incurring quadratic computation and linear memory costs. We present Generative Evolutionary Meta-Solver (GEMS), a surrogate-free framework that replaces explicit populations with a compact set of latent anchors and a single amortized generator. Instead of exhaustively constructing the payoff matrix, GEMS relies on unbiased Monte Carlo rollouts, multiplicative-weights meta-dynamics, and a model-free empirical-Bernstein UCB oracle to adaptively expand the policy set. Best responses are trained within the generator using an advantage-based trust-region objective, eliminating the need to store and train separate actors. We evaluated GEMS in a variety of Two-player and Multi-Player games such as the Deceptive Messages Game, Kuhn Poker and Multi-Particle environment. We find that GEMS is up to ~$\\mathbf{6\\times}$ faster, has $\\mathbf{1.3\\times}$ less memory usage than PSRO, while also reaps higher rewards simultaneously. These results demonstrate that GEMS retains the game theoretic guarantees of PSRO, while overcoming its fundamental inefficiencies, hence enabling scalable multi-agent learning in multiple domains.", "authors": ["Alakh Sharma", "Gaurish Trivedi", "Kartikey Singh Bhandari", "Yash Sinha", "Dhruv Kumar", "Pratik Narang", "Jagat Sesh Challa"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-09-27", "url": "https://arxiv.org/abs/2509.23462", "pdf_url": "https://arxiv.org/pdf/2509.23462v2", "arxiv_id": "2509.23462", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Transactions on Machine Learning Research (2026)", "quality_score": 0.3151}
{"id": "8d027c3ac314a66e4df680ccfab767547c7ad7673374c1acec3313d58b38dfca", "sources": ["arxiv", "semantic_scholar"], "title": "Lifelong Learning with Behavior Consolidation for Vehicle Routing", "abstract": "Recent neural solvers have demonstrated promising performance in learning to solve routing problems. However, existing studies are primarily based on one-off training on one or a set of predefined problem distributions and scales, i.e., tasks. When a new task arises, they typically rely on either zero-shot generalization, which may be poor due to the discrepancies between the new task and the training task(s), or fine-tuning the pretrained solver on the new task, which possibly leads to catastrophic forgetting of knowledge acquired from previous tasks. This paper explores a novel lifelong learning paradigm for neural VRP solvers, where multiple tasks with diverse distributions and scales arise sequentially over time. Solvers are required to effectively and efficiently learn to solve new tasks while maintaining their performance on previously learned tasks. Consequently, a novel framework called Lifelong Learning Router with Behavior Consolidation (LLR-BC) is proposed. LLR-BC consolidates prior knowledge effectively by aligning behaviors of the solver trained on a new task with the buffered ones in a decision-seeking way. To encourage more focus on crucial experiences, LLR-BC assigns greater consolidated weights to decisions with lower confidence. Extensive experiments on capacitated vehicle routing problems and traveling salesman problems demonstrate LLR-BC's effectiveness in training high-performance neural solvers in a lifelong learning setting, addressing the catastrophic forgetting issue, maintaining their plasticity, and improving zero-shot generalization ability.", "authors": ["Jiyuan Pei", "Yi Mei", "Jialin Liu", "Mengjie Zhang", "Xin Yao"], "categories": ["cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-09-26", "url": "https://arxiv.org/abs/2509.21765", "pdf_url": "https://arxiv.org/pdf/2509.21765v4", "arxiv_id": "2509.21765", "doi": "10.48550/arXiv.2509.21765", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.314}
{"id": "c0b01bdc837d0ea3c43cf0a3d646a3023451b12ac9df0fdbfae6e0b0a1109e02", "sources": ["arxiv", "semantic_scholar"], "title": "Intra-Cluster Mixup: An Effective Data Augmentation Technique for Complementary-Label Learning", "abstract": "In this paper, we investigate the challenges of complementary-label learning (CLL), a specialized form of weakly-supervised learning (WSL) where models are trained with labels indicating classes to which instances do not belong, rather than standard ordinary labels. This alternative supervision is appealing because collecting complementary labels is generally cheaper and less labor-intensive. Although most existing research in CLL emphasizes the development of novel loss functions, the potential of data augmentation in this domain remains largely underexplored. In this work, we uncover that the widely-used Mixup data augmentation technique is ineffective when directly applied to CLL. Through in-depth analysis, we identify that the complementary-label noise generated by Mixup negatively impacts the performance of CLL models. We then propose an improved technique called Intra-Cluster Mixup (ICM), which only synthesizes augmented data from nearby examples, to mitigate the noise effect. ICM carries the benefits of encouraging complementary label sharing of nearby examples, and leads to substantial performance improvements across synthetic and real-world labeled datasets. In particular, our wide spectrum of experimental results on both balanced and imbalanced CLL settings justifies the potential of ICM in allying with state-of-the-art CLL algorithms, achieving significant accuracy increases of 30% and 10% on MNIST and CIFAR datasets, respectively.", "authors": ["Tan-Ha Mai", "Hsuan-Tien Lin"], "categories": ["cs.LG", "cs.AI", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2025-09-22", "url": "https://arxiv.org/abs/2509.17971", "pdf_url": "https://arxiv.org/pdf/2509.17971v2", "arxiv_id": "2509.17971", "doi": "10.48550/arXiv.2509.17971", "citation_count": 3, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Transactions on Machine Learning Research, 2026", "quality_score": 0.3094}
{"id": "964fb412be437914c7fa790ee74f680456650d4f4cb6df184347e3b1b007478c", "sources": ["arxiv", "semantic_scholar"], "title": "Fourier Learning Machines: Nonharmonic Fourier-Based Neural Networks for Scientific Machine Learning", "abstract": "We introduce the Fourier Learning Machine (FLM), a neural network (NN) architecture designed to represent a multidimensional nonharmonic Fourier series. The FLM uses a simple feedforward structure with cosine activation functions to learn the frequencies, amplitudes, and phase shifts of the series as trainable parameters. This design allows the model to create a problem-specific spectral basis adaptable to both periodic and nonperiodic functions. Unlike previous Fourier-inspired NN models, the FLM is the first architecture able to represent a multidimensional Fourier series with a complete set of basis functions in separable form, doing so by using a standard Multilayer Perceptron-like architecture. A one-to-one correspondence between the Fourier coefficients and amplitudes and phase-shifts is demonstrated, allowing for the translation between a full, separable basis form and the cosine phase-shifted one. Additionally, we evaluate the performance of FLMs on several scientific computing problems, including benchmark Partial Differential Equations (PDEs) and a family of Optimal Control Problems (OCPs). Computational experiments show that the performance of FLMs is comparable, and often superior, to that of established architectures like SIREN and vanilla feedforward NNs.", "authors": ["Mominul Rubel", "Adam Meyers", "Gabriel Nicolosi"], "categories": ["cs.LG", "math.OC"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2025-09-10", "url": "https://arxiv.org/abs/2509.08759", "pdf_url": "https://arxiv.org/pdf/2509.08759v3", "arxiv_id": "2509.08759", "doi": "10.48550/arXiv.2509.08759", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Transactions on Machine Learning Research, December 2025", "quality_score": 0.2956}
{"id": "925621ad2f75055c0e13214a0c6a4927b140bf9f4df4427650851e72e6d8d93f", "sources": ["arxiv", "semantic_scholar"], "title": "Gaming and Cooperation in Federated Learning: What Can Happen and How to Monitor It", "abstract": "The success of federated learning (FL) ultimately depends on how strategic participants behave under partial observability, yet most formulations still treat FL as a static optimization problem. We instead view FL deployments as governed strategic systems and develop an analytical framework that separates welfare-improving behavior from metric gaming. Within this framework, we introduce indices that quantify manipulability, the price of gaming, and the price of cooperation, and we use them to study how rules, information disclosure, evaluation metrics, and aggregator-switching policies reshape incentives and cooperation patterns. We derive threshold conditions for deterring harmful gaming while preserving benign cooperation, and for triggering auto-switch rules when early-warning indicators become critical. Building on these results, we construct a design toolkit including a governance checklist and a simple audit-budget allocation algorithm with a provable performance guarantee. Simulations across diverse stylized environments and a federated learning case study consistently match the qualitative and quantitative patterns predicted by our framework. Taken together, our results provide design principles and operational guidelines for reducing metric gaming while sustaining stable, high-welfare cooperation in FL platforms.", "authors": ["Dongseok Kim", "Hyoungsun Choi", "Mohamed Jismy Aashik Rasool", "Gisung Oh"], "categories": ["cs.LG", "cs.GT", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2025-09-02", "url": "https://arxiv.org/abs/2509.02391", "pdf_url": "https://arxiv.org/pdf/2509.02391v3", "arxiv_id": "2509.02391", "doi": "10.48550/arXiv.2509.02391", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Transactions on Machine Learning Research, 2026", "quality_score": 0.2865}
{"id": "860d720d28936d99dcf2e9dcf6fc70654482885d2bbdca6703da4aafb23f686b", "sources": ["arxiv", "semantic_scholar"], "title": "Mitigating Catastrophic Forgetting in Continual Learning through Model Growth", "abstract": "Catastrophic forgetting is a significant challenge in continual learning, in which a model loses prior knowledge when it is fine-tuned on new tasks. This problem is particularly critical for large language models (LLMs) undergoing continual learning, as retaining performance across diverse domains is important for their general utility. In this paper, we explore model growth, a promising strategy that leverages smaller models to expedite and structure the training of larger ones for mitigating the catastrophic forgetting problem. Although growth-based pretraining, particularly via transformer stacking, has shown promise in accelerating convergence, its impact on forgetting remains under-explored. Therefore, we evaluate whether growth-based models can retain previously learned capabilities more effectively across a sequence of fine-tuning tasks involving domain knowledge, reasoning, reading comprehension, and bias. Our findings show that both models -- one trained with growth (Stack LLM) and one without (LLM) -- exhibit improvements in domain knowledge. However, reasoning and reading comprehension degrade over time, indicating signs of catastrophic forgetting. Stack LLM consistently shows less degradation, especially in reading comprehension, suggesting enhanced retention capabilities. Interestingly, in bias evaluation, the baseline LLM becomes progressively more neutral with continued fine-tuning, while Stack LLM maintains a steady bias ratio around 60--61\\%. These results indicate that growth-based pretraining may deliver modest improvements in resisting catastrophic forgetting, though trade-offs remain in handling social biases.", "authors": ["Ege Süalp", "Mina Rezaei"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-09-01", "url": "https://arxiv.org/abs/2509.01213", "pdf_url": "https://arxiv.org/pdf/2509.01213v1", "arxiv_id": "2509.01213", "doi": "10.48550/arXiv.2509.01213", "citation_count": 3, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2853}
{"id": "2e50390f1d586dddb134ce20edec8b8e4efa39f1e332949a87e8a6e7b2a72b42", "sources": ["arxiv", "semantic_scholar"], "title": "Online Learning with Multiple Fairness Regularizers via Graph-Structured Feedback", "abstract": "There is an increasing need to enforce multiple, often competing, measures of fairness within automated decision systems. The appropriate weighting of these fairness objectives is typically unknown a priori, may change over time and, in our setting, must be learned adaptively through sequential interactions. In this work, we address this challenge in a bandit setting, where decisions are made with graph-structured feedback.", "authors": ["Quan Zhou", "Jakub Marecek", "Robert Shorten"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-08-19", "url": "https://arxiv.org/abs/2508.14311", "pdf_url": "https://arxiv.org/pdf/2508.14311v2", "arxiv_id": "2508.14311", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Transactions on Machine Learning Research (TMLR), 2026", "quality_score": 0.2704}
{"id": "80cb07b9ce32641a3efa439619a295629fae2fb8c39dfed4162ade58bce5f004", "sources": ["arxiv", "semantic_scholar"], "title": "H2C: Hippocampal Circuit-inspired Continual Learning for Lifelong Trajectory Prediction in Autonomous Driving", "abstract": "Deep learning (DL) has shown state-of-the-art performance in trajectory prediction, which is critical to safe navigation in autonomous driving (AD). However, most DL-based methods suffer from catastrophic forgetting, where adapting to a new distribution may cause significant performance degradation in previously learned ones. Such inability to retain learned knowledge limits their applicability in the real world, where AD systems need to operate across varying scenarios with dynamic distributions. As revealed by neuroscience, the hippocampal circuit plays a crucial role in memory replay, effectively reconstructing learned knowledge based on limited resources. Inspired by this, we propose a hippocampal circuit-inspired continual learning method (H2C) for trajectory prediction across varying scenarios. H2C retains prior knowledge by selectively recalling a small subset of learned samples. First, two complementary strategies are developed to select the subset to represent learned knowledge. Specifically, one strategy maximizes inter-sample diversity to represent the distinctive knowledge, and the other estimates the overall knowledge by equiprobable sampling. Then, H2C updates via a memory replay loss function calculated by these selected samples to retain knowledge while learning new data. Experiments based on various scenarios from the INTERACTION dataset are designed to evaluate H2C. Experimental results show that H2C reduces catastrophic forgetting of DL baselines by 22.71% on average in a task-free manner, without relying on manually informed distributional shifts. The implementation is available at https://github.com/BIT-Jack/H2C-lifelong.", "authors": ["Yunlong Lin", "Zirui Li", "Guodong Du", "Xiaocong Zhao", "Cheng Gong", "Xinwei Wang", "Chao Lu", "Jianwei Gong"], "categories": ["cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-08-02", "url": "https://arxiv.org/abs/2508.01158", "pdf_url": "https://arxiv.org/pdf/2508.01158v2", "arxiv_id": "2508.01158", "doi": "10.48550/arXiv.2508.01158", "citation_count": 2, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/BIT-Jack/H2C-lifelong", "venue": "arXiv.org", "quality_score": 0.3878}
{"id": "68f146317098e49864d32d0f011275318d5bdfd964cf3ffaf3c68e25bc2d992f", "sources": ["arxiv"], "title": "Continual Generalized Category Discovery: Learning and Forgetting from a Bayesian Perspective", "abstract": "Continual Generalized Category Discovery (C-GCD) faces a critical challenge: incrementally learning new classes from unlabeled data streams while preserving knowledge of old classes. Existing methods struggle with catastrophic forgetting, especially when unlabeled data mixes known and novel categories. We address this by analyzing C-GCD's forgetting dynamics through a Bayesian lens, revealing that covariance misalignment between old and new classes drives performance degradation. Building on this insight, we propose Variational Bayes C-GCD (VB-CGCD), a novel framework that integrates variational inference with covariance-aware nearest-class-mean classification. VB-CGCD adaptively aligns class distributions while suppressing pseudo-label noise via stochastic variational updates. Experiments show VB-CGCD surpasses prior art by +15.21% with the overall accuracy in the final session on standard benchmarks. We also introduce a new challenging benchmark with only 10% labeled data and extended online phases, VB-CGCD achieves a 67.86% final accuracy, significantly higher than state-of-the-art (38.55%), demonstrating its robust applicability across diverse scenarios. Code is available at: https://github.com/daihao42/VB-CGCD", "authors": ["Hao Dai", "Jagmohan Chauhan"], "categories": ["cs.LG"], "fields_of_study": [], "published_date": "2025-07-23", "url": "https://arxiv.org/abs/2507.17382", "pdf_url": "https://arxiv.org/pdf/2507.17382v1", "arxiv_id": "2507.17382", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/daihao42/VB-CGCD", "venue": null, "quality_score": 0.283}
{"id": "102542a239d1825abc93b95b2bdd95e6d6a7c647e9e911e610a115e133061965", "sources": ["arxiv", "semantic_scholar"], "title": "Overcoming catastrophic forgetting in neural networks", "abstract": "Catastrophic forgetting is the primary challenge that hinders continual learning, which refers to a neural network ability to sequentially learn multiple tasks while retaining previously acquired knowledge. Elastic Weight Consolidation, a regularization-based approach inspired by synaptic consolidation in biological neural systems, has been used to overcome this problem. In this study prior research is replicated and extended by evaluating EWC in supervised learning settings using the PermutedMNIST and RotatedMNIST benchmarks. Through systematic comparisons with L2 regularization and stochastic gradient descent (SGD) without regularization, we analyze how different approaches balance knowledge retention and adaptability. Our results confirm what was shown in previous research, showing that EWC significantly reduces forgetting compared to naive training while slightly compromising learning efficiency on new tasks. Moreover, we investigate the impact of dropout regularization and varying hyperparameters, offering insights into the generalization of EWC across diverse learning scenarios. These results underscore EWC's potential as a viable solution for lifelong learning in neural networks.", "authors": ["Brandon Shuen Yi Loke", "Filippo Quadri", "Gabriel Vivanco", "Maximilian Casagrande", "Saúl Fenollosa"], "categories": ["cs.LG", "cs.IR"], "fields_of_study": ["Computer Science"], "published_date": "2025-07-14", "url": "https://arxiv.org/abs/2507.10485", "pdf_url": "https://arxiv.org/pdf/2507.10485v1", "arxiv_id": "2507.10485", "doi": "10.48550/arXiv.2507.10485", "citation_count": 160, "influential_citation_count": 15, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.6021}
{"id": "e47a0f50f2fc3ed29368a843d20972839f28346eb9589f15720da6ed00d18e44", "sources": ["arxiv", "semantic_scholar"], "title": "How Weight Resampling and Optimizers Shape the Dynamics of Continual Learning and Forgetting in Neural Networks", "abstract": "Recent work in continual learning has highlighted the beneficial effect of resampling weights in the last layer of a neural network (``zapping\"). Although empirical results demonstrate the effectiveness of this approach, the underlying mechanisms that drive these improvements remain unclear. In this work, we investigate in detail the pattern of learning and forgetting that take place inside a convolutional neural network when trained in challenging settings such as continual learning and few-shot transfer learning, with handwritten characters and natural images. Our experiments show that models that have undergone zapping during training more quickly recover from the shock of transferring to a new domain. Furthermore, to better observe the effect of continual learning in a multi-task setting we measure how each individual task is affected. This shows that, not only zapping, but the choice of optimizer can also deeply affect the dynamics of learning and forgetting, causing complex patterns of synergy/interference between tasks to emerge when the model learns sequentially at transfer time.", "authors": ["Lapo Frati", "Neil Traft", "Jeff Clune", "Nick Cheney"], "categories": ["cs.LG", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2025-07-02", "url": "https://arxiv.org/abs/2507.01559", "pdf_url": "https://arxiv.org/pdf/2507.01559v1", "arxiv_id": "2507.01559", "doi": "10.48550/arXiv.2507.01559", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2154}
{"id": "03fdadd84be14b67b4e959afeac998c72626d098a0e1ab57b5bca6f061c6cdee", "sources": ["arxiv", "semantic_scholar"], "title": "The Importance of Being Lazy: Scaling Limits of Continual Learning", "abstract": "Despite recent efforts, neural networks still struggle to learn in non-stationary environments, and our understanding of catastrophic forgetting (CF) is far from complete. In this work, we perform a systematic study on the impact of model scale and the degree of feature learning in continual learning. We reconcile existing contradictory observations on scale in the literature, by differentiating between lazy and rich training regimes through a variable parameterization of the architecture. We show that increasing model width is only beneficial when it reduces the amount of feature learning, yielding more laziness. Using the framework of dynamical mean field theory, we then study the infinite width dynamics of the model in the feature learning regime and characterize CF, extending prior theoretical results limited to the lazy regime. We study the intricate relationship between feature learning, task non-stationarity, and forgetting, finding that high feature learning is only beneficial with highly similar tasks. We identify a transition modulated by task similarity where the model exits an effectively lazy regime with low forgetting to enter a rich regime with significant forgetting. Finally, our findings reveal that neural networks achieve optimal performance at a critical level of feature learning, which depends on task non-stationarity and transfers across model scales. This work provides a unified perspective on the role of scale and feature learning in continual learning.", "authors": ["Jacopo Graldi", "Alessandro Breccia", "Giulia Lanzillotta", "Thomas Hofmann", "Lorenzo Noci"], "categories": ["cs.LG", "cs.AI", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2025-06-20", "url": "https://arxiv.org/abs/2506.16884", "pdf_url": "https://arxiv.org/pdf/2506.16884v2", "arxiv_id": "2506.16884", "doi": "10.48550/arXiv.2506.16884", "citation_count": 4, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.2017}
{"id": "d627c7ca37ba34f90c9a0f348aacab5831e1e03bf382e25919719546b59be663", "sources": ["arxiv", "semantic_scholar"], "title": "FedOne: Query-Efficient Federated Learning for Black-box Discrete Prompt Learning", "abstract": "Black-Box Discrete Prompt Learning is a prompt-tuning method that optimizes discrete prompts without accessing model parameters or gradients, making the prompt tuning on a cloud-based Large Language Model (LLM) feasible. Adapting federated learning to BDPL could further enhance prompt tuning performance by leveraging data from diverse sources. However, all previous research on federated black-box prompt tuning had neglected the substantial query cost associated with the cloud-based LLM service. To address this gap, we conducted a theoretical analysis of query efficiency within the context of federated black-box prompt tuning. Our findings revealed that degrading FedAvg to activate only one client per round, a strategy we called \\textit{FedOne}, enabled optimal query efficiency in federated black-box prompt learning. Building on this insight, we proposed the FedOne framework, a federated black-box discrete prompt learning method designed to maximize query efficiency when interacting with cloud-based LLMs. We conducted numerical experiments on various aspects of our framework, demonstrating a significant improvement in query efficiency, which aligns with our theoretical results.", "authors": ["Ganyu Wang", "Jinjie Fang", "Maxwell J. Yin", "Bin Gu", "Xi Chen", "Boyu Wang", "Yi Chang", "Charles Ling"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-06-17", "url": "https://arxiv.org/abs/2506.14929", "pdf_url": "https://arxiv.org/pdf/2506.14929v2", "arxiv_id": "2506.14929", "doi": "10.48550/arXiv.2506.14929", "citation_count": 2, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.1982}
{"id": "a4cb0b70b24b402d7546f3c9179fccf00cf87b055777e92d42484473cb17dad8", "sources": ["arxiv", "semantic_scholar"], "title": "On the Similarities of Embeddings in Contrastive Learning", "abstract": "Contrastive learning operates on a simple yet effective principle: Embeddings of positive pairs are pulled together, while those of negative pairs are pushed apart. In this paper, we propose a unified framework for understanding contrastive learning through the lens of cosine similarity, and present two key theoretical insights derived from this framework. First, in full-batch settings, we show that perfect alignment of positive pairs is unattainable when negative-pair similarities fall below a threshold, and this misalignment can be mitigated by incorporating within-view negative pairs into the objective. Second, in mini-batch settings, smaller batch sizes induce stronger separation among negative pairs in the embedding space, i.e., higher variance in their similarities, which in turn degrades the quality of learned representations compared to full-batch settings. To address this, we propose an auxiliary loss that reduces the variance of negative-pair similarities in mini-batch settings. Empirical results show that incorporating the proposed loss improves performance in small-batch settings.", "authors": ["Chungpa Lee", "Sehee Lim", "Kibok Lee", "Jy-yong Sohn"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2025-06-11", "url": "https://arxiv.org/abs/2506.09781", "pdf_url": "https://arxiv.org/pdf/2506.09781v2", "arxiv_id": "2506.09781", "doi": "10.48550/arXiv.2506.09781", "citation_count": 2, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.1914}
{"id": "ee2b08b75c51e46f4b7635e1253d85ca1a18c521036a319853dd7eaeed36be67", "sources": ["arxiv", "semantic_scholar"], "title": "Federated Learning on Stochastic Neural Networks", "abstract": "Federated learning is a machine learning paradigm that leverages edge computing on client devices to optimize models while maintaining user privacy by ensuring that local data remains on the device. However, since all data is collected by clients, federated learning is susceptible to latent noise in local datasets. Factors such as limited measurement capabilities or human errors may introduce inaccuracies in client data. To address this challenge, we propose the use of a stochastic neural network as the local model within the federated learning framework. Stochastic neural networks not only facilitate the estimation of the true underlying states of the data but also enable the quantification of latent noise. We refer to our federated learning approach, which incorporates stochastic neural networks as local models, as Federated stochastic neural networks. We will present numerical experiments demonstrating the performance and effectiveness of our method, particularly in handling non-independent and identically distributed data.", "authors": ["Jingqiao Tang", "Ryan Bausback", "Feng Bao", "Richard Archibald"], "categories": ["cs.LG", "cs.DC"], "fields_of_study": ["Computer Science"], "published_date": "2025-06-09", "url": "https://arxiv.org/abs/2506.08169", "pdf_url": "https://arxiv.org/pdf/2506.08169v1", "arxiv_id": "2506.08169", "doi": "10.48550/arXiv.2506.08169", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1891}
{"id": "a7214f4049d0a30f227516276a52f1b66537e05bea341c926caf35214051b115", "sources": ["arxiv", "semantic_scholar"], "title": "Dynamic Mixture of Progressive Parameter-Efficient Expert Library for Lifelong Robot Learning", "abstract": "A generalist agent must continuously learn and adapt throughout its lifetime, achieving efficient forward transfer while minimizing catastrophic forgetting. Previous work within the dominant pretrain-then-finetune paradigm has explored parameter-efficient fine-tuning for single-task adaptation, effectively steering a frozen pretrained model with a small number of parameters. However, in the context of lifelong learning, these methods rely on the impractical assumption of a test-time task identifier and restrict knowledge sharing among isolated adapters. To address these limitations, we propose Dynamic Mixture of Progressive Parameter-Efficient Expert Library (DMPEL) for lifelong robot learning. DMPEL progressively builds a low-rank expert library and employs a lightweight router to dynamically combine experts into an end-to-end policy, enabling flexible and efficient lifelong forward transfer. Furthermore, by leveraging the modular structure of the fine-tuned parameters, we introduce expert coefficient replay, which guides the router to accurately retrieve frozen experts for previously encountered tasks. This technique mitigates forgetting while being significantly more storage- and computation-efficient than experience replay over the entire policy. Extensive experiments on the lifelong robot learning benchmark LIBERO demonstrate that our framework outperforms state-of-the-art lifelong learning methods in success rates during continual adaptation, while utilizing minimal trainable parameters and storage.", "authors": ["Yuheng Lei", "Sitong Mao", "Shunbo Zhou", "Hongyuan Zhang", "Xuelong Li", "Ping Luo"], "categories": ["cs.LG", "cs.RO"], "fields_of_study": ["Computer Science"], "published_date": "2025-06-06", "url": "https://arxiv.org/abs/2506.05985", "pdf_url": "https://arxiv.org/pdf/2506.05985v3", "arxiv_id": "2506.05985", "doi": "10.48550/arXiv.2506.05985", "citation_count": 4, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/HarryLui98/DMPEL", "venue": "arXiv.org", "quality_score": 0.2869}
{"id": "09a53dbe215f894824ca223cd204aae1ddcbe2f28395cfb2c40bf3b20c00a31d", "sources": ["arxiv", "semantic_scholar"], "title": "Can LLMs Alleviate Catastrophic Forgetting in Graph Continual Learning? A Systematic Study", "abstract": "Nowadays, real-world data, including graph-structure data, often arrives in a streaming manner, which means that learning systems need to continuously acquire new knowledge without forgetting previously learned information. Although substantial existing works attempt to address catastrophic forgetting in graph machine learning, they are all based on training from scratch with streaming data. With the rise of pretrained models, an increasing number of studies have leveraged their strong generalization ability for continual learning. Therefore, in this work, we attempt to answer whether large language models (LLMs) can mitigate catastrophic forgetting in Graph Continual Learning (GCL). We first point out that current experimental setups for GCL have significant flaws, as the evaluation stage may lead to task ID leakage. Then, we evaluate the performance of LLMs in more realistic scenarios and find that even minor modifications can lead to outstanding results. Finally, based on extensive experiments, we propose a simple-yet-effective method, Simple Graph Continual Learning (SimGCL), that surpasses the previous state-of-the-art GNN-based baseline by around 20% under the rehearsal-free constraint. To facilitate reproducibility, we have developed an easy-to-use benchmark LLM4GCL for training and evaluating existing GCL methods. The code is available at: https://github.com/ZhixunLEE/LLM4GCL.", "authors": ["Ziyang Cheng", "Zhixun Li", "Yuhan Li", "Yixin Song", "Kangyi Zhao", "Dawei Cheng", "Jia Li", "Hong Cheng", "Jeffrey Xu Yu"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-05-24", "url": "https://arxiv.org/abs/2505.18697", "pdf_url": "https://arxiv.org/pdf/2505.18697v2", "arxiv_id": "2505.18697", "doi": "10.48550/arXiv.2505.18697", "citation_count": 3, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/ZhixunLEE/LLM4GCL", "venue": "arXiv.org", "quality_score": 0.2639}
{"id": "fd254efb1bf0f3d7f39e441cce8586c91a4c54529e5cd8531c27a281506ab490", "sources": ["arxiv", "semantic_scholar"], "title": "Efficient Training of Neural SDEs Using Stochastic Optimal Control", "abstract": "We present a hierarchical, control theory inspired method for variational inference (VI) for neural stochastic differential equations (SDEs). While VI for neural SDEs is a promising avenue for uncertainty-aware reasoning in time-series, it is computationally challenging due to the iterative nature of maximizing the ELBO. In this work, we propose to decompose the control term into linear and residual non-linear components and derive an optimal control term for linear SDEs, using stochastic optimal control. Modeling the non-linear component by a neural network, we show how to efficiently train neural SDEs without sacrificing their expressive power. Since the linear part of the control term is optimal and does not need to be learned, the training is initialized at a lower cost and we observe faster convergence.", "authors": ["Rembert Daems", "Manfred Opper", "Guillaume Crevecoeur", "Tolga Birdal"], "categories": ["cs.LG", "cs.AI", "math.PR"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2025-05-22", "url": "https://arxiv.org/abs/2505.17150", "pdf_url": "https://arxiv.org/pdf/2505.17150v1", "arxiv_id": "2505.17150", "doi": "10.14428/esann/2025.es2025-182", "citation_count": 3, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "The European Symposium on Artificial Neural Networks", "quality_score": 0.1684}
{"id": "b1bf0e90ff498848e4f7a5e70dd339e95900182d09a2a842edd8bc6bbaf39fdd", "sources": ["arxiv", "semantic_scholar"], "title": "Privacy-Aware Lifelong Learning", "abstract": "Lifelong learning algorithms enable models to incrementally acquire new knowledge without forgetting previously learned information. Contrarily, the field of machine unlearning focuses on explicitly forgetting certain previous knowledge from pretrained models when requested, in order to comply with data privacy regulations on the right-to-be-forgotten. Enabling efficient lifelong learning with the capability to selectively unlearn sensitive information from models presents a critical and largely unaddressed challenge with contradicting objectives. We address this problem from the perspective of simultaneously preventing catastrophic forgetting and allowing forward knowledge transfer during task-incremental learning, while ensuring exact task unlearning and minimizing memory requirements, based on a single neural network model to be adapted. Our proposed solution, privacy-aware lifelong learning (PALL), involves optimization of task-specific sparse subnetworks with parameter sharing within a single architecture. We additionally utilize an episodic memory rehearsal mechanism to facilitate exact unlearning without performance degradations. We empirically demonstrate the scalability of PALL across various architectures in image classification, and provide a state-of-the-art solution that uniquely integrates lifelong learning and privacy-aware unlearning mechanisms for responsible AI applications.", "authors": ["Ozan Özdenizci", "Elmar Rueckert", "Robert Legenstein"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-05-16", "url": "https://arxiv.org/abs/2505.10941", "pdf_url": "https://arxiv.org/pdf/2505.10941v1", "arxiv_id": "2505.10941", "doi": "10.48550/arXiv.2505.10941", "citation_count": 2, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.1616}
{"id": "7966ce1f169d52ae2a78c6512bbeff627fdac9ee7e467615e071e975cec9619d", "sources": ["arxiv", "semantic_scholar"], "title": "Preserving Plasticity in Continual Learning with Adaptive Linearity Injection", "abstract": "Loss of plasticity in deep neural networks is the gradual reduction in a model's capacity to incrementally learn and has been identified as a key obstacle to learning in non-stationary problem settings. Recent work has shown that deep linear networks tend to be resilient towards loss of plasticity. Motivated by this observation, we propose Adaptive Linearization (AdaLin), a general approach that dynamically adapts each neuron's activation function to mitigate plasticity loss. Unlike prior methods that rely on regularization or periodic resets, AdaLin equips every neuron with a learnable parameter and a gating mechanism that injects linearity into the activation function based on its gradient flow. This adaptive modulation ensures sufficient gradient signal and sustains continual learning without introducing additional hyperparameters or requiring explicit task boundaries. When used with conventional activation functions like ReLU, Tanh, and GeLU, we demonstrate that AdaLin can significantly improve performance on standard benchmarks, including Random Label and Permuted MNIST, Random Label and Shuffled CIFAR-10, and Class-Split CIFAR-100. Furthermore, its efficacy is shown in more complex scenarios, such as class-incremental learning on CIFAR-100 with a ResNet-18 backbone, and in mitigating plasticity loss in off-policy reinforcement learning agents. We perform a systematic set of ablations that show that neuron-level adaptation is crucial for good performance and analyze a number of metrics in the network that might be correlated to loss of plasticity.", "authors": ["Seyed Roozbeh Razavi Rohani", "Khashayar Khajavi", "Wesley Chung", "Mo Chen", "Sharan Vaswani"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-05-14", "url": "https://arxiv.org/abs/2505.09486", "pdf_url": "https://arxiv.org/pdf/2505.09486v1", "arxiv_id": "2505.09486", "doi": "10.48550/arXiv.2505.09486", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1593}
{"id": "a83bd4b4abae83fd80c4431cc32f7b8cfe9634e91b4d3b6675921373527417d6", "sources": ["arxiv", "semantic_scholar"], "title": "Bayesian continual learning and forgetting in neural networks", "abstract": "Biological synapses effortlessly balance memory retention and flexibility, yet artificial neural networks still struggle with the extremes of catastrophic forgetting and catastrophic remembering. Here, we introduce Metaplasticity from Synaptic Uncertainty (MESU), a Bayesian framework that updates network parameters according their uncertainty. This approach allows a principled combination of learning and forgetting that ensures that critical knowledge is preserved while unused or outdated information is gradually released. Unlike standard Bayesian approaches -- which risk becoming overly constrained, and popular continual-learning methods that rely on explicit task boundaries, MESU seamlessly adapts to streaming data. It further provides reliable epistemic uncertainty estimates, allowing out-of-distribution detection, the only computational cost being to sample the weights multiple times to provide proper output statistics. Experiments on image-classification benchmarks demonstrate that MESU mitigates catastrophic forgetting, while maintaining plasticity for new tasks. When training 200 sequential permuted MNIST tasks, MESU outperforms established continual learning techniques in terms of accuracy, capability to learn additional tasks, and out-of-distribution data detection. Additionally, due to its non-reliance on task boundaries, MESU outperforms conventional learning techniques on the incremental training of CIFAR-100 tasks consistently in a wide range of scenarios. Our results unify ideas from metaplasticity, Bayesian inference, and Hessian-based regularization, offering a biologically-inspired pathway to robust, perpetual learning.", "authors": ["Djohan Bonnet", "Kellian Cottart", "Tifenn Hirtzlin", "Tarcisius Januel", "Thomas Dalgaty", "Elisa Vianello", "Damien Querlioz"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science", "Medicine"], "published_date": "2025-04-18", "url": "https://arxiv.org/abs/2504.13569", "pdf_url": "https://arxiv.org/pdf/2504.13569v1", "arxiv_id": "2504.13569", "doi": "10.1038/s41467-025-64601-w", "citation_count": 13, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": "Nature Communications", "quality_score": 0.2865}
{"id": "4b11f0cafbf60bf0b47407c4103b255f978ad4c4b2cbff2967ce954f2898a7c5", "sources": ["arxiv", "semantic_scholar"], "title": "Self-Controlled Dynamic Expansion Model for Continual Learning", "abstract": "Continual Learning (CL) epitomizes an advanced training paradigm wherein prior data samples remain inaccessible during the acquisition of new tasks. Numerous investigations have delved into leveraging a pre-trained Vision Transformer (ViT) to enhance model efficacy in continual learning. Nonetheless, these approaches typically utilize a singular, static backbone, which inadequately adapts to novel tasks, particularly when engaging with diverse data domains, due to a substantial number of inactive parameters. This paper addresses this limitation by introducing an innovative Self-Controlled Dynamic Expansion Model (SCDEM), which orchestrates multiple distinct trainable pre-trained ViT backbones to furnish diverse and semantically enriched representations. Specifically, by employing the multi-backbone architecture as a shared module, the proposed SCDEM dynamically generates a new expert with minimal parameters to accommodate a new task. A novel Collaborative Optimization Mechanism (COM) is introduced to synergistically optimize multiple backbones by harnessing prediction signals from historical experts, thereby facilitating new task learning without erasing previously acquired knowledge. Additionally, a novel Feature Distribution Consistency (FDC) approach is proposed to align semantic similarity between previously and currently learned representations through an optimal transport distance-based mechanism, effectively mitigating negative knowledge transfer effects. Furthermore, to alleviate over-regularization challenges, this paper presents a novel Dynamic Layer-Wise Feature Attention Mechanism (DLWFAM) to autonomously determine the penalization intensity on each trainable representation layer. An extensive series of experiments have been conducted to evaluate the proposed methodology's efficacy, with empirical results corroborating that the approach attains state-of-the-art performance.", "authors": ["Runqing Wu", "Kaihui Huang", "Hanyi Zhang", "Fei Ye"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-04-14", "url": "https://arxiv.org/abs/2504.10561", "pdf_url": "https://arxiv.org/pdf/2504.10561v2", "arxiv_id": "2504.10561", "doi": "10.48550/arXiv.2504.10561", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1249}
{"id": "5755c5c44f1c4a0b4222e1c0b4444ec4ea84969a2018bdab0d153b6d52f0af55", "sources": ["arxiv", "semantic_scholar"], "title": "Learning and Improving Backgammon Strategy", "abstract": "A novel approach to learning is presented, combining features of on-line and off-line methods to achieve considerable performance in the task of learning a backgammon value function in a process that exploits the processing power of parallel supercomputers. The off-line methods comprise a set of techniques for parallelizing neural network training and $TD(λ)$ reinforcement learning; here Monte-Carlo ``Rollouts'' are introduced as a massively parallel on-line policy improvement technique which applies resources to the decision points encountered during the search of the game tree to further augment the learned value function estimate. A level of play roughly as good as, or possibly better than, the current champion human and computer backgammon players has been achieved in a short period of learning.", "authors": ["Gregory R. Galperin"], "categories": ["cs.LG", "cs.AI", "cs.NE"], "fields_of_study": ["Computer Science"], "published_date": "2025-04-03", "url": "https://arxiv.org/abs/2504.02221", "pdf_url": "https://arxiv.org/pdf/2504.02221v1", "arxiv_id": "2504.02221", "doi": "10.48550/arXiv.2504.02221", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1123}
{"id": "138ee58c411e657c87d7eb776522b248cdd3fce955424ba9786262a992db1390", "sources": ["arxiv", "semantic_scholar"], "title": "Global Convergence of Continual Learning on Non-IID Data", "abstract": "Continual learning, which aims to learn multiple tasks sequentially, has gained extensive attention. However, most existing work focuses on empirical studies, and the theoretical aspect remains under-explored. Recently, a few investigations have considered the theory of continual learning only for linear regressions, establishes the results based on the strict independent and identically distributed (i.i.d.) assumption and the persistent excitation on the feature data that may be difficult to verify or guarantee in practice. To overcome this fundamental limitation, in this paper, we provide a general and comprehensive theoretical analysis for continual learning of regression models. By utilizing the stochastic Lyapunov function and martingale estimation techniques, we establish the almost sure convergence results of continual learning under a general data condition for the first time. Additionally, without any excitation condition imposed on the data, the convergence rates for the forgetting and regret metrics are provided.", "authors": ["Fei Zhu", "Yujing Liu", "Wenzhuo Liu", "Zhaoxiang Zhang"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-03-24", "url": "https://arxiv.org/abs/2503.18511", "pdf_url": "https://arxiv.org/pdf/2503.18511v1", "arxiv_id": "2503.18511", "doi": "10.48550/arXiv.2503.18511", "citation_count": 3, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1505}
{"id": "4c037600884632076f47305df791ff299a9940eb335895e6610f4086cb17ac86", "sources": ["arxiv", "semantic_scholar"], "title": "Birds look like cars: Adversarial analysis of intrinsically interpretable deep learning", "abstract": "A common belief is that intrinsically interpretable deep learning models ensure a correct, intuitive understanding of their behavior and offer greater robustness against accidental errors or intentional manipulation. However, these beliefs have not been comprehensively verified, and growing evidence casts doubt on them. In this paper, we highlight the risks related to overreliance and susceptibility to adversarial manipulation of these so-called \"intrinsically (aka inherently) interpretable\" models by design. We introduce two strategies for adversarial analysis with prototype manipulation and backdoor attacks against prototype-based networks, and discuss how concept bottleneck models defend against these attacks. Fooling the model's reasoning by exploiting its use of latent prototypes manifests the inherent uninterpretability of deep neural networks, leading to a false sense of security reinforced by a visual confirmation bias. The reported limitations of part-prototype networks put their trustworthiness and applicability into question, motivating further work on the robustness and alignment of (deep) interpretable models.", "authors": ["Hubert Baniecki", "Przemyslaw Biecek"], "categories": ["cs.LG", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2025-03-11", "url": "https://arxiv.org/abs/2503.08636", "pdf_url": "https://arxiv.org/pdf/2503.08636v2", "arxiv_id": "2503.08636", "doi": "10.1007/s10994-025-06896-w", "citation_count": 4, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Machine-mediated learning", "quality_score": 0.1747}
{"id": "6e639f60a69c906f36c3fc1cb2006e42b0f7f8396d1bb81f7e61c749c297fe3e", "sources": ["arxiv", "semantic_scholar"], "title": "A Good Start Matters: Enhancing Continual Learning with Data-Driven Weight Initialization", "abstract": "To adapt to real-world data streams, continual learning (CL) systems must rapidly learn new concepts while preserving and utilizing prior knowledge. When it comes to adding new information to continually-trained deep neural networks (DNNs), classifier weights for newly encountered categories are typically initialized randomly, leading to high initial training loss (spikes) and instability. Consequently, achieving optimal convergence and accuracy requires prolonged training, increasing computational costs. Inspired by Neural Collapse (NC), we propose a weight initialization strategy to improve learning efficiency in CL. In DNNs trained with mean-squared-error, NC gives rise to a Least-Square (LS) classifier in the last layer, whose weights can be analytically derived from learned features. We leverage this LS formulation to initialize classifier weights in a data-driven manner, aligning them with the feature distribution rather than using random initialization. Our method mitigates initial loss spikes and accelerates adaptation to new tasks. We evaluate our approach in large-scale CL settings, demonstrating faster adaptation and improved CL performance.", "authors": ["Md Yousuf Harun", "Christopher Kanan"], "categories": ["cs.LG", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2025-03-09", "url": "https://arxiv.org/abs/2503.06385", "pdf_url": "https://arxiv.org/pdf/2503.06385v2", "arxiv_id": "2503.06385", "doi": "10.48550/arXiv.2503.06385", "citation_count": 2, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1193}
{"id": "5cf2b10c5bea7cb4854bd19497630db3841652cd47e003075ac7f05f6c0f0bac", "sources": ["arxiv", "semantic_scholar"], "title": "Beyond Cosine Decay: On the effectiveness of Infinite Learning Rate Schedule for Continual Pre-training", "abstract": "The ever-growing availability of unlabeled data presents both opportunities and challenges for training artificial intelligence systems. While self-supervised learning (SSL) has emerged as a powerful paradigm for extracting meaningful representations from vast amounts of unlabeled data, existing methods still struggle to adapt to the non-stationary, non-IID nature of real-world data streams without forgetting previously learned knowledge. Recent works have adopted a repeated cosine annealing schedule for large-scale continual pre-training; however, these schedules (1) inherently cause forgetting during the re-warming phase and (2) have not been systematically compared to existing continual SSL methods. In this work, we systematically compare the widely used cosine schedule with the recently proposed infinite learning rate schedule and empirically find the latter to be a more effective alternative. Our extensive empirical evaluation across diverse image and language datasets demonstrates that the infinite learning rate schedule consistently enhances continual pre-training performance compared to a repeated cosine decay without being restricted to a fixed iteration budget. For instance, in a small-scale MAE pre-training setup, it outperforms several strong baselines from the literature. We then scale up our experiments to larger MAE pre-training and autoregressive language model pre-training. Our results show that the infinite learning rate schedule remains effective at scale, surpassing repeated cosine decay for both MAE pre-training and zero-shot LM benchmarks.", "authors": ["Vaibhav Singh", "Paul Janson", "Paria Mehrbod", "Adam Ibrahim", "Irina Rish", "Eugene Belilovsky", "Benjamin Thérien"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-03-04", "url": "https://arxiv.org/abs/2503.02844", "pdf_url": "https://arxiv.org/pdf/2503.02844v3", "arxiv_id": "2503.02844", "doi": "10.48550/arXiv.2503.02844", "citation_count": 5, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1945}
{"id": "a03ab28d3809148a2d2e39d6eebc131391e6343d26a67dc4ca51fe43c5c823c8", "sources": ["arxiv", "semantic_scholar"], "title": "RIZE: Adaptive Regularization for Imitation Learning", "abstract": "We propose a novel Inverse Reinforcement Learning (IRL) method that mitigates the rigidity of fixed reward structures and the limited flexibility of implicit reward regularization. Building on the Maximum Entropy IRL framework, our approach incorporates a squared temporal-difference (TD) regularizer with adaptive targets that evolve dynamically during training, thereby imposing adaptive bounds on recovered rewards and promoting robust decision-making. To capture richer return information, we integrate distributional RL into the learning process. Empirically, our method achieves expert-level performance on complex MuJoCo and Adroit environments, surpassing baseline methods on the Humanoid-v2 task with limited expert demonstrations. Extensive experiments and ablation studies further validate the effectiveness of the approach and provide insights into reward dynamics in imitation learning. Our source code is available at https://github.com/adibka/RIZE.", "authors": ["Adib Karimi", "Mohammad Mehdi Ebadzadeh"], "categories": ["cs.LG", "cs.AI", "cs.RO"], "fields_of_study": ["Computer Science"], "published_date": "2025-02-27", "url": "https://arxiv.org/abs/2502.20089", "pdf_url": "https://arxiv.org/pdf/2502.20089v3", "arxiv_id": "2502.20089", "doi": null, "citation_count": 1, "influential_citation_count": 1, "has_code": true, "code_url": "https://github.com/adibka/RIZE", "venue": "Transactions on Machine Learning Research (11/2025)", "quality_score": 0.1505}
{"id": "a8aef3298d369ec3cd84eff5800403c74479210f739bf13e7c2e8cd0e7b8b09a", "sources": ["arxiv", "semantic_scholar"], "title": "Eidetic Learning: an Efficient and Provable Solution to Catastrophic Forgetting", "abstract": "Catastrophic forgetting -- the phenomenon of a neural network learning a task t1 and losing the ability to perform it after being trained on some other task t2 -- is a long-standing problem for neural networks [McCloskey and Cohen, 1989]. We present a method, Eidetic Learning, that provably solves catastrophic forgetting. A network trained with Eidetic Learning -- here, an EideticNet -- requires no rehearsal or replay. We consider successive discrete tasks and show how at inference time an EideticNet automatically routes new instances without auxiliary task information. An EideticNet bears a family resemblance to the sparsely-gated Mixture-of-Experts layer Shazeer et al. [2016] in that network capacity is partitioned across tasks and the network itself performs data-conditional routing. An EideticNet is easy to implement and train, is efficient, and has time and space complexity linear in the number of parameters. The guarantee of our method holds for normalization layers of modern neural networks during both pre-training and fine-tuning. We show with a variety of network architectures and sets of tasks that EideticNets are immune to forgetting. While the practical benefits of EideticNets are substantial, we believe they can be benefit practitioners and theorists alike. The code for training EideticNets is available at https://github.com/amazon-science/eideticnet-training.", "authors": ["Nicholas Dronen", "Randall Balestriero"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-02-13", "url": "https://arxiv.org/abs/2502.09500", "pdf_url": "https://arxiv.org/pdf/2502.09500v2", "arxiv_id": "2502.09500", "doi": "10.48550/arXiv.2502.09500", "citation_count": 0, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/amazon-science/eideticnet-training", "venue": "arXiv.org", "quality_score": 0.0868}
{"id": "fba79a3a78c214d7b9961510c4b0acb386eddd122b7499bebb21ac21245915a1", "sources": ["arxiv", "semantic_scholar"], "title": "Predicting concentration levels of air pollutants by transfer learning and recurrent neural network", "abstract": "Air pollution (AP) poses a great threat to human health, and people are paying more attention than ever to its prediction. Accurate prediction of AP helps people to plan for their outdoor activities and aids protecting human health. In this paper, long-short term memory (LSTM) recurrent neural networks (RNNs) have been used to predict the future concentration of air pollutants (APS) in Macau. Additionally, meteorological data and data on the concentration of APS have been utilized. Moreover, in Macau, some air quality monitoring stations (AQMSs) have less observed data in quantity, and, at the same time, some AQMSs recorded less observed data of certain types of APS. Therefore, the transfer learning and pre-trained neural networks have been employed to assist AQMSs with less observed data to build a neural network with high prediction accuracy. The experimental sample covers a period longer than 12-year and includes daily measurements from several APS as well as other more classical meteorological values. Records from five stations, four out of them are AQMSs and the remaining one is an automatic weather station, have been prepared from the aforesaid period and eventually underwent to computational intelligence techniques to build and extract a prediction knowledge-based system. As shown by experimentation, LSTM RNNs initialized with transfer learning methods have higher prediction accuracy; it incurred shorter training time than randomly initialized recurrent neural networks.", "authors": ["Iat Hang Fong", "Tengyue Li", "Simon Fong", "Raymond K. Wong", "Antonio J. Tallón-Ballesteros"], "categories": ["cs.LG", "cs.NE", "physics.ao-ph"], "fields_of_study": ["Computer Science", "Physics"], "published_date": "2025-01-30", "url": "https://arxiv.org/abs/2502.01654", "pdf_url": "https://arxiv.org/pdf/2502.01654v1", "arxiv_id": "2502.01654", "doi": "10.1016/j.knosys.2020.105622", "citation_count": 85, "influential_citation_count": 3, "has_code": false, "code_url": null, "venue": "Knowledge-Based Systems", "quality_score": 0.4836}
{"id": "6ae19ac747da297caf8b5c7c3c1bed651226e25e3f65c6b747d7c749c7b6cfe5", "sources": ["arxiv", "semantic_scholar"], "title": "U-Fair: Uncertainty-based Multimodal Multitask Learning for Fairer Depression Detection", "abstract": "Machine learning bias in mental health is becoming an increasingly pertinent challenge. Despite promising efforts indicating that multitask approaches often work better than unitask approaches, there is minimal work investigating the impact of multitask learning on performance and fairness in depression detection nor leveraged it to achieve fairer prediction outcomes. In this work, we undertake a systematic investigation of using a multitask approach to improve performance and fairness for depression detection. We propose a novel gender-based task-reweighting method using uncertainty grounded in how the PHQ-8 questionnaire is structured. Our results indicate that, although a multitask approach improves performance and fairness compared to a unitask approach, the results are not always consistent and we see evidence of negative transfer and a reduction in the Pareto frontier, which is concerning given the high-stake healthcare setting. Our proposed approach of gender-based reweighting with uncertainty improves performance and fairness and alleviates both challenges to a certain extent. Our findings on each PHQ-8 subitem task difficulty are also in agreement with the largest study conducted on the PHQ-8 subitem discrimination capacity, thus providing the very first tangible evidence linking ML findings with large-scale empirical population studies conducted on the PHQ-8.", "authors": ["Jiaee Cheong", "Aditya Bangar", "Sinan Kalkan", "Hatice Gunes"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-01-16", "url": "https://arxiv.org/abs/2501.09687", "pdf_url": "https://arxiv.org/pdf/2501.09687v1", "arxiv_id": "2501.09687", "doi": "10.48550/arXiv.2501.09687", "citation_count": 17, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "Proceedings of Machine Learning Research 2024", "quality_score": 0.3138}
{"id": "fef0b3a2b9108b59024bf79ee189ecfdb9c6cb73f4b3356b5c1e05a268bd4cbf", "sources": ["arxiv", "semantic_scholar"], "title": "An Empirical Analysis of Federated Learning Models Subject to Label-Flipping Adversarial Attack", "abstract": "In this paper, we empirically analyze adversarial attacks on selected federated learning models. The specific learning models considered are Multinominal Logistic Regression (MLR), Support Vector Classifier (SVC), Multilayer Perceptron (MLP), Convolution Neural Network (CNN), %Recurrent Neural Network (RNN), Random Forest, XGBoost, and Long Short-Term Memory (LSTM). For each model, we simulate label-flipping attacks, experimenting extensively with 10 federated clients and 100 federated clients. We vary the percentage of adversarial clients from 10% to 100% and, simultaneously, the percentage of labels flipped by each adversarial client is also varied from 10% to 100%. Among other results, we find that models differ in their inherent robustness to the two vectors in our label-flipping attack, i.e., the percentage of adversarial clients, and the percentage of labels flipped by each adversarial client. We discuss the potential practical implications of our results.", "authors": ["Kunal Bhatnagar", "Sagana Chattanathan", "Angela Dang", "Bhargav Eranki", "Ronnit Rana", "Charan Sridhar", "Siddharth Vedam", "Angie Yao", "Mark Stamp"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-12-24", "url": "https://arxiv.org/abs/2412.18507", "pdf_url": "https://arxiv.org/pdf/2412.18507v1", "arxiv_id": "2412.18507", "doi": "10.48550/arXiv.2412.18507", "citation_count": 2, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1193}
{"id": "ed0cb6cc6cc2f05df477ba7fbcf4871bfd3c3885776ebcc339cb35d20d0f4e75", "sources": ["arxiv", "semantic_scholar"], "title": "Modality-Inconsistent Continual Learning of Multimodal Large Language Models", "abstract": "In this paper, we introduce Modality-Inconsistent Continual Learning (MICL), a new continual learning scenario for Multimodal Large Language Models (MLLMs) that involves tasks with inconsistent modalities (image, audio, or video) and varying task types (captioning or question-answering). Unlike existing vision-only or modality-incremental settings, MICL combines modality and task type shifts, both of which drive catastrophic forgetting. To address these challenges, we propose MoInCL, which employs a Pseudo Targets Generation Module to mitigate forgetting caused by task type shifts in previously seen modalities. It also incorporates Instruction-based Knowledge Distillation to preserve the model's ability to handle previously learned modalities when new ones are introduced. We benchmark MICL using a total of six tasks and conduct experiments to validate the effectiveness of our MoInCL. The experimental results highlight the superiority of MoInCL, showing significant improvements over representative and state-of-the-art continual learning baselines.", "authors": ["Weiguo Pian", "Shijian Deng", "Shentong Mo", "Mingrui Liu", "Yunhui Guo", "Yapeng Tian"], "categories": ["cs.LG", "cs.AI", "cs.CL", "cs.CV", "cs.SD", "eess.AS"], "fields_of_study": ["Computer Science", "Engineering"], "published_date": "2024-12-17", "url": "https://arxiv.org/abs/2412.13050", "pdf_url": "https://arxiv.org/pdf/2412.13050v2", "arxiv_id": "2412.13050", "doi": "10.48550/arXiv.2412.13050", "citation_count": 6, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2113}
{"id": "6650f5ee057d450ca3b997ca6c61d6f5c089ce00521711731b7edbc668eca90c", "sources": ["arxiv", "semantic_scholar"], "title": "Learning to Navigate in Mazes with Novel Layouts using Abstract Top-down Maps", "abstract": "Learning navigation capabilities in different environments has long been one of the major challenges in decision-making. In this work, we focus on zero-shot navigation ability using given abstract $2$-D top-down maps. Like human navigation by reading a paper map, the agent reads the map as an image when navigating in a novel layout, after learning to navigate on a set of training maps. We propose a model-based reinforcement learning approach for this multi-task learning problem, where it jointly learns a hypermodel that takes top-down maps as input and predicts the weights of the transition network. We use the DeepMind Lab environment and customize layouts using generated maps. Our method can adapt better to novel environments in zero-shot and is more robust to noise.", "authors": ["Linfeng Zhao", "Lawson L. S. Wong"], "categories": ["cs.LG", "cs.AI", "cs.RO"], "fields_of_study": ["Computer Science"], "published_date": "2024-12-16", "url": "https://arxiv.org/abs/2412.12024", "pdf_url": "https://arxiv.org/pdf/2412.12024v1", "arxiv_id": "2412.12024", "doi": "10.48550/arXiv.2412.12024", "citation_count": 3, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Journal-ref: Reinforcement Learning Journal, Volume 5, 2024, Pages 2359-2372", "quality_score": 0.1505}
{"id": "9a851e99343d8fa3d43b96b96a7f53f146ebdd64cafe9654a312dcb86d14de6b", "sources": ["arxiv", "semantic_scholar"], "title": "GLL: A Differentiable Graph Learning Layer for Neural Networks", "abstract": "Standard deep learning architectures used for classification generate label predictions with a projection head and softmax activation function. Although successful, these methods fail to leverage the relational information between samples for generating label predictions. In recent works, graph-based learning techniques, namely Laplace learning, have been heuristically combined with neural networks for both supervised and semi-supervised learning (SSL) tasks. However, prior works approximate the gradient of the loss function with respect to the graph learning algorithm or decouple the processes; end-to-end integration with neural networks is not achieved. In this work, we derive backpropagation equations, via the adjoint method, for inclusion of a general family of graph learning layers into a neural network. The resulting method, distinct from graph neural networks, allows us to precisely integrate similarity graph construction and graph Laplacian-based label propagation into a neural network layer, replacing a projection head and softmax activation function for general classification task. Our experimental results demonstrate smooth label transitions across data, improved generalization and robustness to adversarial attacks, and improved training dynamics compared to a standard softmax-based approach.", "authors": ["Jason Brown", "Bohan Chen", "Harris Hardiman-Mostow", "Jeff Calder", "Andrea L. Bertozzi"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2024-12-11", "url": "https://arxiv.org/abs/2412.08016", "pdf_url": "https://arxiv.org/pdf/2412.08016v2", "arxiv_id": "2412.08016", "doi": "10.48550/arXiv.2412.08016", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.0}
{"id": "336f47d06538964e249f6ef1ec40751fb4756b8a10a5d5cb86c846eb9e9b35ea", "sources": ["arxiv", "semantic_scholar"], "title": "Towards Fast Safe Online Reinforcement Learning via Policy Finetuning", "abstract": "The high costs and risks involved in extensive environment interactions hinder the practical application of current online safe reinforcement learning (RL) methods. While offline safe RL addresses this by learning policies from static datasets, the performance therein is usually limited due to reliance on data quality and challenges with out-of-distribution (OOD) actions. Inspired by recent successes in offline-to-online (O2O) RL, it is crucial to explore whether offline safe RL can be leveraged to facilitate faster and safer online policy learning, a direction that has yet to be fully investigated. To fill this gap, we first demonstrate that naively applying existing O2O algorithms from standard RL would not work well in the safe RL setting due to two unique challenges: \\emph{erroneous Q-estimations}, resulted from offline-online objective mismatch and offline cost sparsity, and \\emph{Lagrangian mismatch}, resulted from difficulties in aligning Lagrange multipliers between offline and online policies. To address these challenges, we introduce \\textbf{Marvel}, a novel framework for O2O safe RL, comprising two key components that work in concert: \\emph{Value Pre-Alignment} to align the Q-functions with the underlying truth before online learning, and \\emph{Adaptive PID Control} to effectively adjust the Lagrange multipliers during online finetuning. Extensive experiments demonstrate that Marvel significantly outperforms existing baselines in both reward maximization and safety constraint satisfaction. By introducing the first policy-finetuning based framework for O2O safe RL, which is compatible with many offline and online safe RL methods, our work has the great potential to advance the field towards more efficient and practical safe RL solutions.", "authors": ["Keru Chen", "Honghao Wei", "Zhigang Deng", "Sen Lin"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-12-05", "url": "https://arxiv.org/abs/2412.04426", "pdf_url": "https://arxiv.org/pdf/2412.04426v4", "arxiv_id": "2412.04426", "doi": null, "citation_count": 5, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Transactions on Machine Learning Research (TMLR), 2026", "quality_score": 0.1945}
{"id": "d1f352e14340829e04ef51162c4fa6a002b7fd0366724f7f86df8884bea95043", "sources": ["arxiv", "semantic_scholar"], "title": "Robust Offline Reinforcement Learning with Linearly Structured f-Divergence Regularization", "abstract": "The Robust Regularized Markov Decision Process (RRMDP) is proposed to learn policies robust to dynamics shifts by adding regularization to the transition dynamics in the value function. Existing methods mostly use unstructured regularization, potentially leading to conservative policies under unrealistic transitions. To address this limitation, we propose a novel framework, the $d$-rectangular linear RRMDP ($d$-RRMDP), which introduces latent structures into both transition kernels and regularization. We focus on offline reinforcement learning, where an agent learns policies from a precollected dataset in the nominal environment. We develop the Robust Regularized Pessimistic Value Iteration (R2PVI) algorithm that employs linear function approximation for robust policy learning in $d$-RRMDPs with $f$-divergence based regularization terms on transition kernels. We provide instance-dependent upper bounds on the suboptimality gap of R2PVI policies, demonstrating that these bounds are influenced by how well the dataset covers state-action spaces visited by the optimal robust policy under robustly admissible transitions. We establish information-theoretic lower bounds to verify that our algorithm is near-optimal. Finally, numerical experiments validate that R2PVI learns robust policies and exhibits superior computational efficiency compared to baseline methods.", "authors": ["Cheng Tang", "Zhishuai Liu", "Pan Xu"], "categories": ["cs.LG", "cs.AI", "cs.RO", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2024-11-27", "url": "https://arxiv.org/abs/2411.18612", "pdf_url": "https://arxiv.org/pdf/2411.18612v2", "arxiv_id": "2411.18612", "doi": "10.48550/arXiv.2411.18612", "citation_count": 6, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.2113}
{"id": "c3fcd4e5924e106b357a800e9defcc8bd682502f41c3f5c502c2564159630e5e", "sources": ["arxiv", "semantic_scholar"], "title": "Learning Explainable Treatment Policies with Clinician-Informed Representations: A Practical Approach", "abstract": "Digital health interventions (DHIs) and remote patient monitoring (RPM) have shown great potential in improving chronic disease management through personalized care. However, barriers like limited efficacy and workload concerns hinder adoption of existing DHIs; while limited sample sizes and lack of interpretability limit the effectiveness and adoption of purely black-box algorithmic DHIs. In this paper, we address these challenges by developing a pipeline for learning explainable treatment policies for RPM-enabled DHIs. We apply our approach in the real-world setting of RPM using a DHI to improve glycemic control of youth with type 1 diabetes. Our main contribution is to reveal the importance of clinical domain knowledge in developing state and action representations for effective, efficient, and interpretable targeting policies. We observe that policies learned from clinician-informed representations are significantly more efficacious and efficient than policies learned from black-box representations. This work emphasizes the importance of collaboration between ML researchers and clinicians for developing effective DHIs in the real world.", "authors": ["Johannes O. Ferstad", "Emily B. Fox", "David Scheinker", "Ramesh Johari"], "categories": ["cs.LG", "cs.AI", "stat.AP", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2024-11-26", "url": "https://arxiv.org/abs/2411.17570", "pdf_url": "https://arxiv.org/pdf/2411.17570v1", "arxiv_id": "2411.17570", "doi": "10.48550/arXiv.2411.17570", "citation_count": 1, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/jferstad/ml4h-explainable-policies", "venue": "Proceedings of the 4th Machine Learning for Health Symposium, PMLR 259:325-349, 2025", "quality_score": 0.0753}
{"id": "80bc76755004ee91762cff2835acf7bea5bdf12d6a1897b4206fb80a1e79e63d", "sources": ["arxiv", "semantic_scholar"], "title": "Slowing Down Forgetting in Continual Learning", "abstract": "A common challenge in continual learning (CL) is catastrophic forgetting, where the performance on old tasks drops after new, additional tasks are learned. In this paper, we propose a novel framework called ReCL to slow down forgetting in CL. Our framework exploits an implicit bias of gradient-based neural networks due to which these converge to margin maximization points. Such convergence points allow us to reconstruct old data from previous tasks, which we then combine with the current training data. Our framework is flexible and can be applied on top of existing, state-of-the-art CL methods. We further demonstrate the performance gain from our framework across a large series of experiments, including two challenging CL scenarios (class incremental and domain incremental learning), different datasets (MNIST, CIFAR10, TinyImagenet), and different network architectures. Across all experiments, we find large performance gains through ReCL. To the best of our knowledge, our framework is the first to address catastrophic forgetting by leveraging models in CL as their own memory buffers.", "authors": ["Pascal Janetzky", "Tobias Schlagenhauf", "Stefan Feuerriegel"], "categories": ["cs.LG", "cs.AI", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2024-11-11", "url": "https://arxiv.org/abs/2411.06916", "pdf_url": "https://arxiv.org/pdf/2411.06916v2", "arxiv_id": "2411.06916", "doi": "10.48550/arXiv.2411.06916", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.0}
{"id": "52ad4e434a517294fe71e828e65bfb28d5226553ddfdd3154819f2e6cebea43f", "sources": ["arxiv", "semantic_scholar"], "title": "Return Augmented Decision Transformer for Off-Dynamics Reinforcement Learning", "abstract": "We study offline off-dynamics reinforcement learning (RL) to utilize data from an easily accessible source domain to enhance policy learning in a target domain with limited data. Our approach centers on return-conditioned supervised learning (RCSL), particularly focusing on Decision Transformer (DT) type frameworks, which can predict actions conditioned on desired return guidance and complete trajectory history. Previous works address the dynamics shift problem by augmenting the reward in the trajectory from the source domain to match the optimal trajectory in the target domain. However, this strategy can not be directly applicable in RCSL owing to (1) the unique form of the RCSL policy class, which explicitly depends on the return, and (2) the absence of a straightforward representation of the optimal trajectory distribution. We propose the Return Augmented (REAG) method for DT type frameworks, where we augment the return in the source domain by aligning its distribution with that in the target domain. We provide the theoretical analysis demonstrating that the RCSL policy learned from REAG achieves the same level of suboptimality as would be obtained without a dynamics shift. We introduce two practical implementations REAG$_\\text{Dara}^{*}$ and REAG$_\\text{MV}^{*}$ respectively. Thorough experiments on D4RL datasets and various DT-type baselines demonstrate that our methods consistently enhance the performance of DT type frameworks in off-dynamics RL.", "authors": ["Ruhan Wang", "Yu Yang", "Zhishuai Liu", "Dongruo Zhou", "Pan Xu"], "categories": ["cs.LG", "cs.AI", "cs.RO", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2024-10-30", "url": "https://arxiv.org/abs/2410.23450", "pdf_url": "https://arxiv.org/pdf/2410.23450v2", "arxiv_id": "2410.23450", "doi": "10.48550/arXiv.2410.23450", "citation_count": 14, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Transactions on Machine Learning Research, 2026", "quality_score": 0.294}
{"id": "d244f818afce974bf87dbe7d9fffb5c86a95b0950adbd27675118e23aea1a544", "sources": ["arxiv", "semantic_scholar"], "title": "The Effects of Multi-Task Learning on ReLU Neural Network Functions", "abstract": "This paper studies the properties of solutions to multi-task shallow ReLU neural network learning problems, wherein the network is trained to fit a dataset with minimal sum of squared weights. Remarkably, the solutions learned for each individual task resemble those obtained by solving a kernel regression problem, revealing a novel connection between neural networks and kernel methods. It is known that single-task neural network learning problems are equivalent to a minimum norm interpolation problem in a non-Hilbertian Banach space, and that the solutions of such problems are generally non-unique. In contrast, we prove that the solutions to univariate-input, multi-task neural network interpolation problems are almost always unique, and coincide with the solution to a minimum-norm interpolation problem in a Sobolev (Reproducing Kernel) Hilbert Space. We also demonstrate a similar phenomenon in the multivariate-input case; specifically, we show that neural network learning problems with large numbers of tasks are approximately equivalent to an $\\ell^2$ (Hilbert space) minimization problem over a fixed kernel determined by the optimal neurons.", "authors": ["Julia Nakhleh", "Joseph Shenouda", "Robert D. Nowak"], "categories": ["stat.ML", "cs.LG"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2024-10-29", "url": "https://arxiv.org/abs/2410.21696", "pdf_url": "https://arxiv.org/pdf/2410.21696v4", "arxiv_id": "2410.21696", "doi": "10.48550/arXiv.2410.21696", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.0753}
{"id": "8012bfeca2d752a9b972c82020ce91042e5aacca1df4a4e23e40402d53bb8028", "sources": ["arxiv", "semantic_scholar"], "title": "Improving Multimodal Large Language Models Using Continual Learning", "abstract": "Generative large language models (LLMs) exhibit impressive capabilities, which can be further augmented by integrating a pre-trained vision model into the original LLM to create a multimodal LLM (MLLM). However, this integration often significantly decreases performance on natural language understanding and generation tasks, compared to the original LLM. This study investigates this issue using the LLaVA MLLM, treating the integration as a continual learning problem. We evaluate five continual learning methods to mitigate forgetting and identify a technique that enhances visual understanding while minimizing linguistic performance loss. Our approach reduces linguistic performance degradation by up to 15% over the LLaVA recipe, while maintaining high multimodal accuracy. We also demonstrate the robustness of our method through continual learning on a sequence of vision-language tasks, effectively preserving linguistic skills while acquiring new multimodal capabilities. Project webpage: https://shikhar-srivastava.github.io/cl-for-improving-mllms", "authors": ["Shikhar Srivastava", "Md Yousuf Harun", "Robik Shrestha", "Christopher Kanan"], "categories": ["cs.CL", "cs.CV", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-10-25", "url": "https://arxiv.org/abs/2410.19925", "pdf_url": "https://arxiv.org/pdf/2410.19925v2", "arxiv_id": "2410.19925", "doi": "10.48550/arXiv.2410.19925", "citation_count": 4, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1747}
{"id": "2043f6e7b9a8be39d290778da6b9ea8dec44b9228f20671bb63ea5788455a711", "sources": ["arxiv", "semantic_scholar"], "title": "SNAP: Stopping Catastrophic Forgetting in Hebbian Learning with Sigmoidal Neuronal Adaptive Plasticity", "abstract": "Artificial Neural Networks (ANNs) suffer from catastrophic forgetting, where the learning of new tasks causes the catastrophic forgetting of old tasks. Existing Machine Learning (ML) algorithms, including those using Stochastic Gradient Descent (SGD) and Hebbian Learning typically update their weights linearly with experience i.e., independently of their current strength. This contrasts with biological neurons, which at intermediate strengths are very plastic, but consolidate with Long-Term Potentiation (LTP) once they reach a certain strength. We hypothesize this mechanism might help mitigate catastrophic forgetting. We introduce Sigmoidal Neuronal Adaptive Plasticity (SNAP) an artificial approximation to Long-Term Potentiation for ANNs by having the weights follow a sigmoidal growth behaviour allowing the weights to consolidate and stabilize when they reach sufficiently large or small values. We then compare SNAP to linear weight growth and exponential weight growth and see that SNAP completely prevents the forgetting of previous tasks for Hebbian Learning but not for SGD-base learning.", "authors": ["Tianyi Xu", "Patrick Zheng", "Shiyan Liu", "Sicheng Lyu", "Isabeau Prémont-Schwarz"], "categories": ["cs.NE", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-10-20", "url": "https://arxiv.org/abs/2410.15318", "pdf_url": "https://arxiv.org/pdf/2410.15318v1", "arxiv_id": "2410.15318", "doi": "10.48550/arXiv.2410.15318", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.0}
{"id": "7e9282a7a056712a6f1547fc1a5959dab7ad1eb48d0391e99692c3534698b2eb", "sources": ["arxiv", "semantic_scholar"], "title": "Continual Deep Reinforcement Learning to Prevent Catastrophic Forgetting in Jamming Mitigation", "abstract": "Deep Reinforcement Learning (DRL) has been highly effective in learning from and adapting to RF environments and thus detecting and mitigating jamming effects to facilitate reliable wireless communications. However, traditional DRL methods are susceptible to catastrophic forgetting (namely forgetting old tasks when learning new ones), especially in dynamic wireless environments where jammer patterns change over time. This paper considers an anti-jamming system and addresses the challenge of catastrophic forgetting in DRL applied to jammer detection and mitigation. First, we demonstrate the impact of catastrophic forgetting in DRL when applied to jammer detection and mitigation tasks, where the network forgets previously learned jammer patterns while adapting to new ones. This catastrophic interference undermines the effectiveness of the system, particularly in scenarios where the environment is non-stationary. We present a method that enables the network to retain knowledge of old jammer patterns while learning to handle new ones. Our approach substantially reduces catastrophic forgetting, allowing the anti-jamming system to learn new tasks without compromising its ability to perform previously learned tasks effectively. Furthermore, we introduce a systematic methodology for sequentially learning tasks in the anti-jamming framework. By leveraging continual DRL techniques based on PackNet, we achieve superior anti-jamming performance compared to standard DRL methods. Our proposed approach not only addresses catastrophic forgetting but also enhances the adaptability and robustness of the system in dynamic jamming environments. We demonstrate the efficacy of our method in preserving knowledge of past jammer patterns, learning new tasks efficiently, and achieving superior anti-jamming performance compared to traditional DRL approaches.", "authors": ["Kemal Davaslioglu", "Sastry Kompella", "Tugba Erpek", "Yalin E. Sagduyu"], "categories": ["cs.LG", "cs.AI", "cs.NI"], "fields_of_study": ["Computer Science"], "published_date": "2024-10-14", "url": "https://arxiv.org/abs/2410.10521", "pdf_url": "https://arxiv.org/pdf/2410.10521v1", "arxiv_id": "2410.10521", "doi": "10.1109/MILCOM61039.2024.10773861", "citation_count": 7, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "IEEE Military Communications Conference", "quality_score": 0.2258}
{"id": "d6200f974bff103076f9972960728197bbf44cdf6d5796bf7027abcd8b8d0570", "sources": ["arxiv", "semantic_scholar"], "title": "Metalic: Meta-Learning In-Context with Protein Language Models", "abstract": "Predicting the biophysical and functional properties of proteins is essential for in silico protein design. Machine learning has emerged as a promising technique for such prediction tasks. However, the relative scarcity of in vitro annotations means that these models often have little, or no, specific data on the desired fitness prediction task. As a result of limited data, protein language models (PLMs) are typically trained on general protein sequence modeling tasks, and then fine-tuned, or applied zero-shot, to protein fitness prediction. When no task data is available, the models make strong assumptions about the correlation between the protein sequence likelihood and fitness scores. In contrast, we propose meta-learning over a distribution of standard fitness prediction tasks, and demonstrate positive transfer to unseen fitness prediction tasks. Our method, called Metalic (Meta-Learning In-Context), uses in-context learning and fine-tuning, when data is available, to adapt to new tasks. Crucially, fine-tuning enables considerable generalization, even though it is not accounted for during meta-training. Our fine-tuned models achieve strong results with 18 times fewer parameters than state-of-the-art models. Moreover, our method sets a new state-of-the-art in low-data settings on ProteinGym, an established fitness-prediction benchmark. Due to data scarcity, we believe meta-learning will play a pivotal role in advancing protein engineering.", "authors": ["Jacob Beck", "Shikha Surana", "Manus McAuliffe", "Oliver Bent", "Thomas D. Barrett", "Juan Jose Garau Luis", "Paul Duckworth"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-10-10", "url": "https://arxiv.org/abs/2410.08355", "pdf_url": "https://arxiv.org/pdf/2410.08355v3", "arxiv_id": "2410.08355", "doi": "10.48550/arXiv.2410.08355", "citation_count": 4, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/instadeepai/metalic", "venue": "International Conference on Learning Representations", "quality_score": 0.1747}
{"id": "79659c2c098850941dda1357e7297579a3dd8988bcc48a7df745726018b67441", "sources": ["arxiv", "semantic_scholar"], "title": "Scalable Mechanistic Neural Networks for Differential Equations and Machine Learning", "abstract": "We propose Scalable Mechanistic Neural Network (S-MNN), an enhanced neural network framework designed for scientific machine learning applications involving long temporal sequences. By reformulating the original Mechanistic Neural Network (MNN) (Pervez et al., 2024), we reduce the computational time and space complexities from cubic and quadratic with respect to the sequence length, respectively, to linear. This significant improvement enables efficient modeling of long-term dynamics without sacrificing accuracy or interpretability. Extensive experiments demonstrate that S-MNN matches the original MNN in precision while substantially reducing computational resources. Consequently, S-MNN can drop-in replace the original MNN in applications, providing a practical and efficient tool for integrating mechanistic bottlenecks into neural network models of complex dynamical systems. Source code is available at https://github.com/IST-DASLab/ScalableMNN.", "authors": ["Jiale Chen", "Dingling Yao", "Adeel Pervez", "Dan Alistarh", "Francesco Locatello"], "categories": ["cs.LG", "math.NA"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2024-10-08", "url": "https://arxiv.org/abs/2410.06074", "pdf_url": "https://arxiv.org/pdf/2410.06074v3", "arxiv_id": "2410.06074", "doi": "10.48550/arXiv.2410.06074", "citation_count": 4, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/IST-DASLab/ScalableMNN", "venue": "International Conference on Learning Representations", "quality_score": 0.1747}
{"id": "279d0bc6be060ae753ae9adf8f74d3f12a2fe4c2df3fb9430845fd7e5e761dbb", "sources": ["arxiv", "semantic_scholar"], "title": "Recent Advances of Multimodal Continual Learning: A Comprehensive Survey", "abstract": "Continual learning (CL) aims to empower machine learning models to learn continually from new data, while building upon previously acquired knowledge without forgetting. As models have evolved from small to large pre-trained architectures, and from supporting unimodal to multimodal data, multimodal continual learning (MMCL) methods have recently emerged. The primary complexity of MMCL is that it extends beyond a simple stacking of unimodal CL methods. Such straightforward approaches often suffer from multimodal catastrophic forgetting, yielding unsatisfactory performance. In addition, MMCL introduces new challenges that unimodal CL methods fail to adequately address, including modality imbalance, complex modality interaction, high computational costs, and degradation of pre-trained zero-shot capability of multimodal backbones. In this work, we present the first comprehensive survey on MMCL. We provide essential background knowledge and MMCL settings, as well as a structured taxonomy of MMCL methods. We categorize MMCL methods into four categories, i.e., regularization-based, architecture-based, replay-based, and prompt-based methods, explaining their methodologies and highlighting their key innovations. Additionally, to prompt further research in this field, we summarize open MMCL datasets and benchmarks, provide an in-depth discussion, and discuss several promising future directions. We have also created a GitHub repository for indexing relevant MMCL papers and open resources available at https://github.com/LucyDYu/Awesome-Multimodal-Continual-Learning.", "authors": ["Dianzhi Yu", "Xinni Zhang", "Yankai Chen", "Aiwei Liu", "Yifei Zhang", "Philip S. Yu", "Irwin King"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Medicine", "Computer Science"], "published_date": "2024-10-07", "url": "https://arxiv.org/abs/2410.05352", "pdf_url": "https://arxiv.org/pdf/2410.05352v3", "arxiv_id": "2410.05352", "doi": "10.48550/arXiv.2410.05352", "citation_count": 45, "influential_citation_count": 1, "has_code": true, "code_url": "https://github.com/LucyDYu/Awesome-Multimodal-Continual-Learning", "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.4157}
{"id": "03a355a6a4ec6179d93ce97995cc020129ae233fbc6cdec422c34d34a3236d7b", "sources": ["arxiv", "semantic_scholar"], "title": "Optimal Protocols for Continual Learning via Statistical Physics and Control Theory", "abstract": "Artificial neural networks often struggle with catastrophic forgetting when learning multiple tasks sequentially, as training on new tasks degrades the performance on previously learned tasks. Recent theoretical work has addressed this issue by analysing learning curves in synthetic frameworks under predefined training protocols. However, these protocols relied on heuristics and lacked a solid theoretical foundation assessing their optimality. In this paper, we fill this gap by combining exact equations for training dynamics, derived using statistical physics techniques, with optimal control methods. We apply this approach to teacher-student models for continual learning and multi-task problems, obtaining a theory for task-selection protocols maximising performance while minimising forgetting. Our theoretical analysis offers non-trivial yet interpretable strategies for mitigating catastrophic forgetting, shedding light on how optimal learning protocols modulate established effects, such as the influence of task similarity on forgetting. Finally, we validate our theoretical findings with experiments on real-world data.", "authors": ["Francesco Mori", "Stefano Sarao Mannelli", "Francesca Mignacco"], "categories": ["cs.LG", "cond-mat.dis-nn", "cond-mat.stat-mech"], "fields_of_study": ["Computer Science", "Physics"], "published_date": "2024-09-26", "url": "https://arxiv.org/abs/2409.18061", "pdf_url": "https://arxiv.org/pdf/2409.18061v3", "arxiv_id": "2409.18061", "doi": "10.1088/1742-5468/adf296", "citation_count": 15, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.301}
{"id": "eabdeafb5e974afa235ed22714c1ad2ee61cb6dd60a8afaf6f5622d4d2165997", "sources": ["arxiv", "semantic_scholar"], "title": "Patch-Based Contrastive Learning and Memory Consolidation for Online Unsupervised Continual Learning", "abstract": "We focus on a relatively unexplored learning paradigm known as {\\em Online Unsupervised Continual Learning} (O-UCL), where an agent receives a non-stationary, unlabeled data stream and progressively learns to identify an increasing number of classes. This paradigm is designed to model real-world applications where encountering novelty is the norm, such as exploring a terrain with several unknown and time-varying entities. Unlike prior work in unsupervised, continual, or online learning, O-UCL combines all three areas into a single challenging and realistic learning paradigm. In this setting, agents are frequently evaluated and must aim to maintain the best possible representation at any point of the data stream, rather than at the end of pre-specified offline tasks. The proposed approach, called \\textbf{P}atch-based \\textbf{C}ontrastive learning and \\textbf{M}emory \\textbf{C}onsolidation (PCMC), builds a compositional understanding of data by identifying and clustering patch-level features. Embeddings for these patch-level features are extracted with an encoder trained via patch-based contrastive learning. PCMC incorporates new data into its distribution while avoiding catastrophic forgetting, and it consolidates memory examples during ``sleep\" periods. We evaluate PCMC's performance on streams created from the ImageNet and Places365 datasets. Additionally, we explore various versions of the PCMC algorithm and compare its performance against several existing methods and simple baselines.", "authors": ["Cameron Taylor", "Vassilis Vassiliades", "Constantine Dovrolis"], "categories": ["cs.LG", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2024-09-24", "url": "https://arxiv.org/abs/2409.16391", "pdf_url": "https://arxiv.org/pdf/2409.16391v1", "arxiv_id": "2409.16391", "doi": "10.48550/arXiv.2409.16391", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.0753}
{"id": "1e352c84ec288d33aebe43af05b954e283899ecfd570f19d67607b8c3a41c1da", "sources": ["arxiv", "semantic_scholar"], "title": "A Contrastive Symmetric Forward-Forward Algorithm (SFFA) for Continual Learning Tasks", "abstract": "The so-called Forward-Forward Algorithm (FFA) has recently gained momentum as an alternative to the conventional back-propagation algorithm for neural network learning, yielding competitive performance across various modeling tasks. By replacing the backward pass of gradient back-propagation with two contrastive forward passes, the FFA avoids several shortcomings undergone by its predecessor (e.g., vanishing/exploding gradient) by enabling layer-wise training heuristics. In classification tasks, this contrastive method has been proven to effectively create a latent sparse representation of the input data, ultimately favoring discriminability. However, FFA exhibits an inherent asymmetric gradient behavior due to an imbalanced loss function between positive and negative data, adversely impacting on the model's generalization capabilities and leading to an accuracy degradation. To address this issue, this work proposes the Symmetric Forward-Forward Algorithm (SFFA), a novel modification of the original FFA which partitions each layer into positive and negative neurons. This allows the local fitness function to be defined as the ratio between the activation of positive neurons and the overall layer activity, resulting in a symmetric loss landscape during the training phase. To evaluate the enhanced convergence of our method, we conduct several experiments using multiple image classification benchmarks, comparing the accuracy of models trained with SFFA to those trained with its FFA counterpart. As a byproduct of this reformulation, we explore the advantages of using a layer-wise training algorithm for Continual Learning (CL) tasks. The specialization of neurons and the sparsity of their activations induced by layer-wise training algorithms enable efficient CL strategies that incorporate new knowledge (classes) into the neural network, while preventing catastrophic forgetting of previously...", "authors": ["Erik B. Terres-Escudero", "Javier Del Ser", "Pablo Garcia Bringas"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-09-11", "url": "https://arxiv.org/abs/2409.07387", "pdf_url": "https://arxiv.org/pdf/2409.07387v2", "arxiv_id": "2409.07387", "doi": "10.48550/arXiv.2409.07387", "citation_count": 2, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.1193}
{"id": "da3176728c186e3d4604f510c332aac0f53dfb431b383d93ea9a0b9e76192fd0", "sources": ["arxiv", "semantic_scholar"], "title": "Buffer-based Gradient Projection for Continual Federated Learning", "abstract": "Continual Federated Learning (CFL) is essential for enabling real-world applications where multiple decentralized clients adaptively learn from continuous data streams. A significant challenge in CFL is mitigating catastrophic forgetting, where models lose previously acquired knowledge when learning new information. Existing approaches often face difficulties due to the constraints of device storage capacities and the heterogeneous nature of data distributions among clients. While some CFL algorithms have addressed these challenges, they frequently rely on unrealistic assumptions about the availability of task boundaries (i.e., knowing when new tasks begin). To address these limitations, we introduce Fed-A-GEM, a federated adaptation of the A-GEM method (Chaudhry et al., 2019), which employs a buffer-based gradient projection approach. Fed-A-GEM alleviates catastrophic forgetting by leveraging local buffer samples and aggregated buffer gradients, thus preserving knowledge across multiple clients. Our method is combined with existing CFL techniques, enhancing their performance in the CFL context. Our experiments on standard benchmarks show consistent performance improvements across diverse scenarios. For example, in a task-incremental learning scenario using the CIFAR-100 dataset, our method can increase the accuracy by up to 27%. Our code is available at https://github.com/shenghongdai/Fed-A-GEM.", "authors": ["Shenghong Dai", "Jy-yong Sohn", "Yicong Chen", "S M Iftekharul Alam", "Ravikumar Balakrishnan", "Suman Banerjee", "Nageen Himayat", "Kangwook Lee"], "categories": ["cs.LG", "cs.DC"], "fields_of_study": ["Computer Science"], "published_date": "2024-09-03", "url": "https://arxiv.org/abs/2409.01585", "pdf_url": "https://arxiv.org/pdf/2409.01585v1", "arxiv_id": "2409.01585", "doi": "10.48550/arXiv.2409.01585", "citation_count": 4, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/shenghongdai/Fed-A-GEM", "venue": null, "quality_score": 0.1747}
{"id": "f5f33f268f5c99164da5c472f7b0e369b5e4f69b344d5dbf79418d4ca21e2902", "sources": ["arxiv", "semantic_scholar"], "title": "Continual learning with the neural tangent ensemble", "abstract": "A natural strategy for continual learning is to weigh a Bayesian ensemble of fixed functions. This suggests that if a (single) neural network could be interpreted as an ensemble, one could design effective algorithms that learn without forgetting. To realize this possibility, we observe that a neural network classifier with N parameters can be interpreted as a weighted ensemble of N classifiers, and that in the lazy regime limit these classifiers are fixed throughout learning. We call these classifiers the neural tangent experts and show they output valid probability distributions over the labels. We then derive the likelihood and posterior probability of each expert given past data. Surprisingly, the posterior updates for these experts are equivalent to a scaled and projected form of stochastic gradient descent (SGD) over the network weights. Away from the lazy regime, networks can be seen as ensembles of adaptive experts which improve over time. These results offer a new interpretation of neural networks as Bayesian ensembles of experts, providing a principled framework for understanding and mitigating catastrophic forgetting in continual learning settings.", "authors": ["Ari S. Benjamin", "Christian Pehle", "Kyle Daruwalla"], "categories": ["cs.LG", "cs.NE"], "fields_of_study": ["Computer Science"], "published_date": "2024-08-30", "url": "https://arxiv.org/abs/2408.17394", "pdf_url": "https://arxiv.org/pdf/2408.17394v2", "arxiv_id": "2408.17394", "doi": "10.48550/arXiv.2408.17394", "citation_count": 3, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Neural Information Processing Systems", "quality_score": 0.1505}
{"id": "86e72946f2dd7bfa094f7329d135f0c4be66cbb44ff1346f445d8d8ab2d6ac66", "sources": ["arxiv", "semantic_scholar"], "title": "Learning Multi-Index Models with Neural Networks via Mean-Field Langevin Dynamics", "abstract": "We study the problem of learning multi-index models in high-dimensions using a two-layer neural network trained with the mean-field Langevin algorithm. Under mild distributional assumptions on the data, we characterize the effective dimension $d_{\\mathrm{eff}}$ that controls both sample and computational complexity by utilizing the adaptivity of neural networks to latent low-dimensional structures. When the data exhibit such a structure, $d_{\\mathrm{eff}}$ can be significantly smaller than the ambient dimension. We prove that the sample complexity grows almost linearly with $d_{\\mathrm{eff}}$, bypassing the limitations of the information and generative exponents that appeared in recent analyses of gradient-based feature learning. On the other hand, the computational complexity may inevitably grow exponentially with $d_{\\mathrm{eff}}$ in the worst-case scenario. Motivated by improving computational complexity, we take the first steps towards polynomial time convergence of the mean-field Langevin algorithm by investigating a setting where the weights are constrained to be on a compact manifold with positive Ricci curvature, such as the hypersphere. There, we study assumptions under which polynomial time convergence is achievable, whereas similar assumptions in the Euclidean setting lead to exponential time complexity.", "authors": ["Alireza Mousavi-Hosseini", "Denny Wu", "Murat A. Erdogdu"], "categories": ["stat.ML", "cs.LG"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2024-08-14", "url": "https://arxiv.org/abs/2408.07254", "pdf_url": "https://arxiv.org/pdf/2408.07254v2", "arxiv_id": "2408.07254", "doi": "10.48550/arXiv.2408.07254", "citation_count": 13, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.2865}
{"id": "3f5265d68735dc4ec97a6e975ef93031500d986c5cacf4de40bbac2cb0a7cdf4", "sources": ["arxiv", "semantic_scholar"], "title": "Learning to Learn without Forgetting using Attention", "abstract": "Continual learning (CL) refers to the ability to continually learn over time by accommodating new knowledge while retaining previously learned experience. While this concept is inherent in human learning, current machine learning methods are highly prone to overwrite previously learned patterns and thus forget past experience. Instead, model parameters should be updated selectively and carefully, avoiding unnecessary forgetting while optimally leveraging previously learned patterns to accelerate future learning. Since hand-crafting effective update mechanisms is difficult, we propose meta-learning a transformer-based optimizer to enhance CL. This meta-learned optimizer uses attention to learn the complex relationships between model parameters across a stream of tasks, and is designed to generate effective weight updates for the current task while preventing catastrophic forgetting on previously encountered tasks. Evaluations on benchmark datasets like SplitMNIST, RotatedMNIST, and SplitCIFAR-100 affirm the efficacy of the proposed approach in terms of both forward and backward transfer, even on small sets of labeled data, highlighting the advantages of integrating a meta-learned optimizer within the continual learning framework.", "authors": ["Anna Vettoruzzo", "Joaquin Vanschoren", "Mohamed-Rafik Bouguelia", "Thorsteinn Rögnvaldsson"], "categories": ["cs.LG", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2024-08-06", "url": "https://arxiv.org/abs/2408.03219", "pdf_url": "https://arxiv.org/pdf/2408.03219v2", "arxiv_id": "2408.03219", "doi": "10.48550/arXiv.2408.03219", "citation_count": 2, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.1193}
{"id": "cd080e33193dc5b9ca8a3e10ffb3580a10a8712527b3283c6c07e74b4d8a81bd", "sources": ["arxiv", "semantic_scholar"], "title": "Diffusion Augmented Agents: A Framework for Efficient Exploration and Transfer Learning", "abstract": "We introduce Diffusion Augmented Agents (DAAG), a novel framework that leverages large language models, vision language models, and diffusion models to improve sample efficiency and transfer learning in reinforcement learning for embodied agents. DAAG hindsight relabels the agent's past experience by using diffusion models to transform videos in a temporally and geometrically consistent way to align with target instructions with a technique we call Hindsight Experience Augmentation. A large language model orchestrates this autonomous process without requiring human supervision, making it well-suited for lifelong learning scenarios. The framework reduces the amount of reward-labeled data needed to 1) finetune a vision language model that acts as a reward detector, and 2) train RL agents on new tasks. We demonstrate the sample efficiency gains of DAAG in simulated robotics environments involving manipulation and navigation. Our results show that DAAG improves learning of reward detectors, transferring past experience, and acquiring new tasks - key abilities for developing efficient lifelong learning agents. Supplementary material and visualizations are available on our website https://sites.google.com/view/diffusion-augmented-agents/", "authors": ["Norman Di Palo", "Leonard Hasenclever", "Jan Humplik", "Arunkumar Byravan"], "categories": ["cs.LG", "cs.AI", "cs.RO"], "fields_of_study": ["Computer Science"], "published_date": "2024-07-30", "url": "https://arxiv.org/abs/2407.20798", "pdf_url": "https://arxiv.org/pdf/2407.20798v1", "arxiv_id": "2407.20798", "doi": "10.48550/arXiv.2407.20798", "citation_count": 5, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.1945}
{"id": "63a8a658e4435a28415dacc4b78d1091e74508f12a3b348539cb583cc6d788c7", "sources": ["arxiv", "semantic_scholar"], "title": "Gradient Boosting Reinforcement Learning", "abstract": "We present Gradient Boosting Reinforcement Learning (GBRL), a framework that adapts the strengths of gradient boosting trees (GBT) to reinforcement learning (RL) tasks. While neural networks (NNs) have become the de facto choice for RL, they face significant challenges with structured and categorical features and tend to generalize poorly to out-of-distribution samples. These are challenges for which GBTs have traditionally excelled in supervised learning. However, GBT's application in RL has been limited. The design of traditional GBT libraries is optimized for static datasets with fixed labels, making them incompatible with RL's dynamic nature, where both state distributions and reward signals evolve during training. GBRL overcomes this limitation by continuously interleaving tree construction with environment interaction. Through extensive experiments, we demonstrate that GBRL outperforms NNs in domains with structured observations and categorical features while maintaining competitive performance on standard continuous control benchmarks. Like its supervised learning counterpart, GBRL demonstrates superior robustness to out-of-distribution samples and better handles irregular state-action relationships.", "authors": ["Benjamin Fuhrer", "Chen Tessler", "Gal Dalal"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-07-11", "url": "https://arxiv.org/abs/2407.08250", "pdf_url": "https://arxiv.org/pdf/2407.08250v2", "arxiv_id": "2407.08250", "doi": "10.48550/arXiv.2407.08250", "citation_count": 5, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.1945}
{"id": "eea226d772d1c320a9074c69e411018fb2842a97de61e6f1e71a266d85b333ff", "sources": ["arxiv", "semantic_scholar"], "title": "How to Leverage Predictive Uncertainty Estimates for Reducing Catastrophic Forgetting in Online Continual Learning", "abstract": "Many real-world applications require machine-learning models to be able to deal with non-stationary data distributions and thus learn autonomously over an extended period of time, often in an online setting. One of the main challenges in this scenario is the so-called catastrophic forgetting (CF) for which the learning model tends to focus on the most recent tasks while experiencing predictive degradation on older ones. In the online setting, the most effective solutions employ a fixed-size memory buffer to store old samples used for replay when training on new tasks. Many approaches have been presented to tackle this problem. However, it is not clear how predictive uncertainty information for memory management can be leveraged in the most effective manner and conflicting strategies are proposed to populate the memory. Are the easiest-to-forget or the easiest-to-remember samples more effective in combating CF? Starting from the intuition that predictive uncertainty provides an idea of the samples' location in the decision space, this work presents an in-depth analysis of different uncertainty estimates and strategies for populating the memory. The investigation provides a better understanding of the characteristics data points should have for alleviating CF. Then, we propose an alternative method for estimating predictive uncertainty via the generalised variance induced by the negative log-likelihood. Finally, we demonstrate that the use of predictive uncertainty measures helps in reducing CF in different settings.", "authors": ["Giuseppe Serra", "Ben Werner", "Florian Buettner"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-07-10", "url": "https://arxiv.org/abs/2407.07668", "pdf_url": "https://arxiv.org/pdf/2407.07668v3", "arxiv_id": "2407.07668", "doi": "10.48550/arXiv.2407.07668", "citation_count": 7, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2258}
{"id": "c9b3bdf3aa7df4821a997ff6b58f9263de964e63456938063a6e4c5bb5706250", "sources": ["arxiv", "semantic_scholar"], "title": "The impact of model size on catastrophic forgetting in Online Continual Learning", "abstract": "This study investigates the impact of model size on Online Continual Learning performance, with a focus on catastrophic forgetting. Employing ResNet architectures of varying sizes, the research examines how network depth and width affect model performance in class-incremental learning using the SplitCIFAR-10 dataset. Key findings reveal that larger models do not guarantee better Continual Learning performance; in fact, they often struggle more in adapting to new tasks, particularly in online settings. These results challenge the notion that larger models inherently mitigate catastrophic forgetting, highlighting the nuanced relationship between model size and Continual Learning efficacy. This study contributes to a deeper understanding of model scalability and its practical implications in Continual Learning scenarios.", "authors": ["Eunhae Lee"], "categories": ["cs.LG", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-28", "url": "https://arxiv.org/abs/2407.00176", "pdf_url": "https://arxiv.org/pdf/2407.00176v1", "arxiv_id": "2407.00176", "doi": "10.48550/arXiv.2407.00176", "citation_count": 3, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1505}
{"id": "c2158d623790a66666a21e9630c134ba184756d90dba986b2bdcae7300285dd0", "sources": ["arxiv", "semantic_scholar"], "title": "Retrospective Feature Estimation for Continual Learning", "abstract": "The intrinsic capability to continuously learn a changing data stream is a desideratum of deep neural networks (DNNs). However, current DNNs suffer from catastrophic forgetting, which interferes with remembering past knowledge. To mitigate this issue, existing Continual Learning (CL) approaches often retain exemplars for replay, regularize learning, or allocate dedicated capacity for new tasks. This paper investigates an unexplored direction for CL called Retrospective Feature Estimation (RFE). RFE learns to reverse feature changes by aligning the features from the current trained DNN backward to the feature space of the old task, where performing predictions is easier. This retrospective process utilizes a chain of small feature mapping networks called retrospector modules. Empirical experiments on several CL benchmarks, including CIFAR10, CIFAR100, and Tiny ImageNet, demonstrate the effectiveness and potential of this novel CL direction compared to existing representative CL methods, motivating further research into retrospective mechanisms as a principled alternative for mitigating catastrophic forgetting in CL. Code is available at: https://github.com/mail-research/retrospective-feature-estimation.", "authors": ["Nghia D. Nguyen", "Hieu Trung Nguyen", "Ang Li", "Hoang Pham", "Viet Anh Nguyen", "Khoa D. Doan"], "categories": ["cs.LG", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-25", "url": "https://arxiv.org/abs/2406.17381", "pdf_url": "https://arxiv.org/pdf/2406.17381v2", "arxiv_id": "2406.17381", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/mail-research/retrospective-feature-estimation", "venue": null, "quality_score": 0.0}
{"id": "c49f55f2262da9c4078d9ae7f69474e34804c669fba2e9771ccc0c0746a82271", "sources": ["arxiv", "semantic_scholar"], "title": "Learning Temporal Distances: Contrastive Successor Features Can Provide a Metric Structure for Decision-Making", "abstract": "Temporal distances lie at the heart of many algorithms for planning, control, and reinforcement learning that involve reaching goals, allowing one to estimate the transit time between two states. However, prior attempts to define such temporal distances in stochastic settings have been stymied by an important limitation: these prior approaches do not satisfy the triangle inequality. This is not merely a definitional concern, but translates to an inability to generalize and find shortest paths. In this paper, we build on prior work in contrastive learning and quasimetrics to show how successor features learned by contrastive learning (after a change of variables) form a temporal distance that does satisfy the triangle inequality, even in stochastic settings. Importantly, this temporal distance is computationally efficient to estimate, even in high-dimensional and stochastic settings. Experiments in controlled settings and benchmark suites demonstrate that an RL algorithm based on these new temporal distances exhibits combinatorial generalization (i.e., \"stitching\") and can sometimes learn more quickly than prior methods, including those based on quasimetrics.", "authors": ["Vivek Myers", "Chongyi Zheng", "Anca Dragan", "Sergey Levine", "Benjamin Eysenbach"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-24", "url": "https://arxiv.org/abs/2406.17098", "pdf_url": "https://arxiv.org/pdf/2406.17098v2", "arxiv_id": "2406.17098", "doi": "10.48550/arXiv.2406.17098", "citation_count": 48, "influential_citation_count": 4, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.4225}
{"id": "b7202610fc64bca15aeebf80c4d3b217ba154ed2dbfb06cf512d711320975f98", "sources": ["arxiv", "semantic_scholar"], "title": "Towards evolution of Deep Neural Networks through contrastive Self-Supervised learning", "abstract": "Deep Neural Networks (DNNs) have been successfully applied to a wide range of problems. However, two main limitations are commonly pointed out. The first one is that they require long time to design. The other is that they heavily rely on labelled data, which can sometimes be costly and hard to obtain. In order to address the first problem, neuroevolution has been proved to be a plausible option to automate the design of DNNs. As for the second problem, self-supervised learning has been used to leverage unlabelled data to learn representations. Our goal is to study how neuroevolution can help self-supervised learning to bridge the gap to supervised learning in terms of performance. In this work, we propose a framework that is able to evolve deep neural networks using self-supervised learning. Our results on the CIFAR-10 dataset show that it is possible to evolve adequate neural networks while reducing the reliance on labelled data. Moreover, an analysis to the structure of the evolved networks suggests that the amount of labelled data fed to them has less effect on the structure of networks that learned via self-supervised learning, when compared to individuals that relied on supervised learning.", "authors": ["Adriano Vinhas", "João Correia", "Penousal Machado"], "categories": ["cs.NE", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-20", "url": "https://arxiv.org/abs/2406.14525", "pdf_url": "https://arxiv.org/pdf/2406.14525v1", "arxiv_id": "2406.14525", "doi": "10.1109/CEC60901.2024.10611910", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "IEEE Congress on Evolutionary Computation", "quality_score": 0.0}
{"id": "1282ee9230196185d57ad379f5af61cea0fe36d3fa8e3b2518f0cabbf611f7fa", "sources": ["arxiv", "semantic_scholar"], "title": "A deep cut into Split Federated Self-supervised Learning", "abstract": "Collaborative self-supervised learning has recently become feasible in highly distributed environments by dividing the network layers between client devices and a central server. However, state-of-the-art methods, such as MocoSFL, are optimized for network division at the initial layers, which decreases the protection of the client data and increases communication overhead. In this paper, we demonstrate that splitting depth is crucial for maintaining privacy and communication efficiency in distributed training. We also show that MocoSFL suffers from a catastrophic quality deterioration for the minimal communication overhead. As a remedy, we introduce Momentum-Aligned contrastive Split Federated Learning (MonAcoSFL), which aligns online and momentum client models during training procedure. Consequently, we achieve state-of-the-art accuracy while significantly reducing the communication overhead, making MonAcoSFL more practical in real-world scenarios.", "authors": ["Marcin Przewięźlikowski", "Marcin Osial", "Bartosz Zieliński", "Marek Śmieja"], "categories": ["cs.LG", "cs.AI", "cs.DC"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-12", "url": "https://arxiv.org/abs/2406.08267", "pdf_url": "https://arxiv.org/pdf/2406.08267v2", "arxiv_id": "2406.08267", "doi": "10.1007/978-3-031-70344-7_26", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Machine Learning and Knowledge Discovery in Databases. Research Track. ECML PKDD 2024. Lecture Notes in Computer Science, vol 14942. Springer, Cham", "quality_score": 0.0753}
{"id": "a37d6512c43498f6a950a777150edff34ec729491088d581cdfd93bb7d43413c", "sources": ["arxiv", "semantic_scholar"], "title": "Coprocessor Actor Critic: A Model-Based Reinforcement Learning Approach For Adaptive Brain Stimulation", "abstract": "Adaptive brain stimulation can treat neurological conditions such as Parkinson's disease and post-stroke motor deficits by influencing abnormal neural activity. Because of patient heterogeneity, each patient requires a unique stimulation policy to achieve optimal neural responses. Model-free reinforcement learning (MFRL) holds promise in learning effective policies for a variety of similar control tasks, but is limited in domains like brain stimulation by a need for numerous costly environment interactions. In this work we introduce Coprocessor Actor Critic, a novel, model-based reinforcement learning (MBRL) approach for learning neural coprocessor policies for brain stimulation. Our key insight is that coprocessor policy learning is a combination of learning how to act optimally in the world and learning how to induce optimal actions in the world through stimulation of an injured brain. We show that our approach overcomes the limitations of traditional MFRL methods in terms of sample efficiency and task success and outperforms baseline MBRL approaches in a neurologically realistic model of an injured brain.", "authors": ["Michelle Pan", "Mariah Schrum", "Vivek Myers", "Erdem Bıyık", "Anca Dragan"], "categories": ["cs.LG", "cs.AI", "cs.HC"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-10", "url": "https://arxiv.org/abs/2406.06714", "pdf_url": "https://arxiv.org/pdf/2406.06714v2", "arxiv_id": "2406.06714", "doi": "10.48550/arXiv.2406.06714", "citation_count": 2, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.1193}
{"id": "6f12abfd48c58ccca7a7af987f101dabebefb3f073cc8060a69cbf023c951358", "sources": ["arxiv"], "title": "Stabilizing Extreme Q-learning by Maclaurin Expansion", "abstract": "In offline reinforcement learning, in-sample learning methods have been widely used to prevent performance degradation caused by evaluating out-of-distribution actions from the dataset. Extreme Q-learning (XQL) employs a loss function based on the assumption that Bellman error follows a Gumbel distribution, enabling it to model the soft optimal value function in an in-sample manner. It has demonstrated strong performance in both offline and online reinforcement learning settings. However, issues remain, such as the instability caused by the exponential term in the loss function and the risk of the error distribution deviating from the Gumbel distribution. Therefore, we propose Maclaurin Expanded Extreme Q-learning to enhance stability. In this method, applying Maclaurin expansion to the loss function in XQL enhances stability against large errors. This approach involves adjusting the modeled value function between the value function under the behavior policy and the soft optimal value function, thus achieving a trade-off between stability and optimality depending on the order of expansion. It also enables adjustment of the error distribution assumption from a normal distribution to a Gumbel distribution. Our method significantly stabilizes learning in online RL tasks from DM Control, where XQL was previously unstable. Additionally, it improves performance in several offline RL tasks from D4RL.", "authors": ["Motoki Omura", "Takayuki Osa", "Yusuke Mukuta", "Tatsuya Harada"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": [], "published_date": "2024-06-07", "url": "https://arxiv.org/abs/2406.04896", "pdf_url": "https://arxiv.org/pdf/2406.04896v2", "arxiv_id": "2406.04896", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Reinforcement Learning Journal, 2024, Volume 3, pages 1427-1440", "quality_score": 0.0}
{"id": "2acc90a5fd42c173cb9b5ac4e2533a78ae6c237d136f060f906d20b37b647c26", "sources": ["arxiv", "semantic_scholar"], "title": "Pretraining Decision Transformers with Reward Prediction for In-Context Multi-task Structured Bandit Learning", "abstract": "We study learning to learn for the multi-task structured bandit problem where the goal is to learn a near-optimal algorithm that minimizes cumulative regret. The tasks share a common structure and an algorithm should exploit the shared structure to minimize the cumulative regret for an unseen but related test task. We use a transformer as a decision-making algorithm to learn this shared structure from data collected by a demonstrator on a set of training task instances. Our objective is to devise a training procedure such that the transformer will learn to outperform the demonstrator's learning algorithm on unseen test task instances. Prior work on pretraining decision transformers either requires privileged information like access to optimal arms or cannot outperform the demonstrator. Going beyond these approaches, we introduce a pre-training approach that trains a transformer network to learn a near-optimal policy in-context. This approach leverages the shared structure across tasks, does not require access to optimal actions, and can outperform the demonstrator. We validate these claims over a wide variety of structured bandit problems to show that our proposed solution is general and can quickly identify expected rewards on unseen test tasks to support effective exploration.", "authors": ["Subhojyoti Mukherjee", "Josiah P. Hanna", "Qiaomin Xie", "Robert Nowak"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-07", "url": "https://arxiv.org/abs/2406.05064", "pdf_url": "https://arxiv.org/pdf/2406.05064v3", "arxiv_id": "2406.05064", "doi": "10.48550/arXiv.2406.05064", "citation_count": 9, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.25}
{"id": "aeff4a784fc5dbe7d8ac19028cc723cc06496bf013b668e42885b0950fa69fac", "sources": ["arxiv", "semantic_scholar"], "title": "Large Brain Model for Learning Generic Representations with Tremendous EEG Data in BCI", "abstract": "The current electroencephalogram (EEG) based deep learning models are typically designed for specific datasets and applications in brain-computer interaction (BCI), limiting the scale of the models and thus diminishing their perceptual capabilities and generalizability. Recently, Large Language Models (LLMs) have achieved unprecedented success in text processing, prompting us to explore the capabilities of Large EEG Models (LEMs). We hope that LEMs can break through the limitations of different task types of EEG datasets, and obtain universal perceptual capabilities of EEG signals through unsupervised pre-training. Then the models can be fine-tuned for different downstream tasks. However, compared to text data, the volume of EEG datasets is generally small and the format varies widely. For example, there can be mismatched numbers of electrodes, unequal length data samples, varied task designs, and low signal-to-noise ratio. To overcome these challenges, we propose a unified foundation model for EEG called Large Brain Model (LaBraM). LaBraM enables cross-dataset learning by segmenting the EEG signals into EEG channel patches. Vector-quantized neural spectrum prediction is used to train a semantically rich neural tokenizer that encodes continuous raw EEG channel patches into compact neural codes. We then pre-train neural Transformers by predicting the original neural codes for the masked EEG channel patches. The LaBraMs were pre-trained on about 2,500 hours of various types of EEG signals from around 20 datasets and validated on multiple different types of downstream tasks. Experiments on abnormal detection, event type classification, emotion recognition, and gait prediction show that our LaBraM outperforms all compared SOTA methods in their respective fields. Our code is available at https://github.com/935963004/LaBraM.", "authors": ["Wei-Bang Jiang", "Li-Ming Zhao", "Bao-Liang Lu"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-05-29", "url": "https://arxiv.org/abs/2405.18765", "pdf_url": "https://arxiv.org/pdf/2405.18765v1", "arxiv_id": "2405.18765", "doi": "10.48550/arXiv.2405.18765", "citation_count": 350, "influential_citation_count": 98, "has_code": true, "code_url": "https://github.com/935963004/LaBraM", "venue": "International Conference on Learning Representations", "quality_score": 0.9978}
{"id": "d6fc803e4ef44406743c018f3b945123785fc5c10b2d1728e1f1aa7da08abc02", "sources": ["arxiv", "semantic_scholar"], "title": "Understanding Forgetting in Continual Learning with Linear Regression", "abstract": "Continual learning, focused on sequentially learning multiple tasks, has gained significant attention recently. Despite the tremendous progress made in the past, the theoretical understanding, especially factors contributing to catastrophic forgetting, remains relatively unexplored. In this paper, we provide a general theoretical analysis of forgetting in the linear regression model via Stochastic Gradient Descent (SGD) applicable to both underparameterized and overparameterized regimes. Our theoretical framework reveals some interesting insights into the intricate relationship between task sequence and algorithmic parameters, an aspect not fully captured in previous studies due to their restrictive assumptions. Specifically, we demonstrate that, given a sufficiently large data size, the arrangement of tasks in a sequence, where tasks with larger eigenvalues in their population data covariance matrices are trained later, tends to result in increased forgetting. Additionally, our findings highlight that an appropriate choice of step size will help mitigate forgetting in both underparameterized and overparameterized settings. To validate our theoretical analysis, we conducted simulation experiments on both linear regression models and Deep Neural Networks (DNNs). Results from these simulations substantiate our theoretical findings.", "authors": ["Meng Ding", "Kaiyi Ji", "Di Wang", "Jinhui Xu"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-05-27", "url": "https://arxiv.org/abs/2405.17583", "pdf_url": "https://arxiv.org/pdf/2405.17583v1", "arxiv_id": "2405.17583", "doi": "10.48550/arXiv.2405.17583", "citation_count": 26, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.3578}
{"id": "073333e61d36792cb98e0d68805f9b96a0b635d41ba3741777113fb91898f961", "sources": ["arxiv", "semantic_scholar"], "title": "Densely Distilling Cumulative Knowledge for Continual Learning", "abstract": "Continual learning, involving sequential training on diverse tasks, often faces catastrophic forgetting. While knowledge distillation-based approaches exhibit notable success in preventing forgetting, we pinpoint a limitation in their ability to distill the cumulative knowledge of all the previous tasks. To remedy this, we propose Dense Knowledge Distillation (DKD). DKD uses a task pool to track the model's capabilities. It partitions the output logits of the model into dense groups, each corresponding to a task in the task pool. It then distills all tasks' knowledge using all groups. However, using all the groups can be computationally expensive, we also suggest random group selection in each optimization step. Moreover, we propose an adaptive weighting scheme, which balances the learning of new classes and the retention of old classes, based on the count and similarity of the classes. Our DKD outperforms recent state-of-the-art baselines across diverse benchmarks and scenarios. Empirical analysis underscores DKD's ability to enhance model stability, promote flatter minima for improved generalization, and remains robust across various memory budgets and task orders. Moreover, it seamlessly integrates with other CL methods to boost performance and proves versatile in offline scenarios like model compression.", "authors": ["Zenglin Shi", "Pei Liu", "Tong Su", "Yunpeng Wu", "Kuien Liu", "Yu Song", "Meng Wang"], "categories": ["cs.LG", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2024-05-16", "url": "https://arxiv.org/abs/2405.09820", "pdf_url": "https://arxiv.org/pdf/2405.09820v1", "arxiv_id": "2405.09820", "doi": "10.48550/arXiv.2405.09820", "citation_count": 3, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1505}
{"id": "21000fec970cd7b001f7b29cac5020d6f3d665bbec733d0e4709418022d1a046", "sources": ["arxiv", "semantic_scholar"], "title": "SSFL: Discovering Sparse Unified Subnetworks at Initialization for Efficient Federated Learning", "abstract": "In this work, we propose Salient Sparse Federated Learning (SSFL), a streamlined approach for sparse federated learning with efficient communication. SSFL identifies a sparse subnetwork prior to training, leveraging parameter saliency scores computed separately on local client data in non-IID scenarios, and then aggregated, to determine a global mask. Only the sparse model weights are trained and communicated each round between the clients and the server. On standard benchmarks including CIFAR-10, CIFAR-100, and Tiny-ImageNet, SSFL consistently improves the accuracy sparsity trade off, achieving more than 20\\% relative error reduction on CIFAR-10 compared to the strongest sparse baseline, while reducing communication costs by $2 \\times$ relative to dense FL. Finally, in a real-world federated learning deployment, SSFL delivers over $2.3 \\times$ faster communication time, underscoring its practical efficiency.", "authors": ["Riyasat Ohib", "Bishal Thapaliya", "Gintare Karolina Dziugaite", "Jingyu Liu", "Vince Calhoun", "Sergey Plis"], "categories": ["cs.LG", "cs.AI", "cs.DC"], "fields_of_study": ["Computer Science"], "published_date": "2024-05-15", "url": "https://arxiv.org/abs/2405.09037", "pdf_url": "https://arxiv.org/pdf/2405.09037v2", "arxiv_id": "2405.09037", "doi": null, "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Transactions on Machine Learning Research, 2026", "quality_score": 0.0753}
{"id": "5b90e9de69a4bb0d18aa4e68456c20fdc81e6f5c05207d6b667af60878d066b6", "sources": ["arxiv", "semantic_scholar"], "title": "Continual Learning in the Presence of Repetition", "abstract": "Continual learning (CL) provides a framework for training models in ever-evolving environments. Although re-occurrence of previously seen objects or tasks is common in real-world problems, the concept of repetition in the data stream is not often considered in standard benchmarks for CL. Unlike with the rehearsal mechanism in buffer-based strategies, where sample repetition is controlled by the strategy, repetition in the data stream naturally stems from the environment. This report provides a summary of the CLVision challenge at CVPR 2023, which focused on the topic of repetition in class-incremental learning. The report initially outlines the challenge objective and then describes three solutions proposed by finalist teams that aim to effectively exploit the repetition in the stream to learn continually. The experimental results from the challenge highlight the effectiveness of ensemble-based solutions that employ multiple versions of similar modules, each trained on different but overlapping subsets of classes. This report underscores the transformative potential of taking a different perspective in CL by employing repetition in the data stream to foster innovative strategy design.", "authors": ["Hamed Hemati", "Lorenzo Pellegrini", "Xiaotian Duan", "Zixuan Zhao", "Fangfang Xia", "Marc Masana", "Benedikt Tscheschner", "Eduardo Veas", "Yuxiang Zheng", "Shiji Zhao", "Shao-Yuan Li", "Sheng-Jun Huang", "Vincenzo Lomonaco", "Gido M. van de Ven"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science", "Medicine"], "published_date": "2024-05-07", "url": "https://arxiv.org/abs/2405.04101", "pdf_url": "https://arxiv.org/pdf/2405.04101v2", "arxiv_id": "2405.04101", "doi": "10.1016/j.neunet.2024.106920", "citation_count": 9, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Neural Networks", "quality_score": 0.25}
{"id": "91a56f869887d7071dcda9ccf9734ddab74499664cb404ce1a1eb1e47c3c4e37", "sources": ["arxiv", "semantic_scholar"], "title": "Representation Learning of Daily Movement Data Using Text Encoders", "abstract": "Time-series representation learning is a key area of research for remote healthcare monitoring applications. In this work, we focus on a dataset of recordings of in-home activity from people living with Dementia. We design a representation learning method based on converting activity to text strings that can be encoded using a language model fine-tuned to transform data from the same participants within a $30$-day window to similar embeddings in the vector space. This allows for clustering and vector searching over participants and days, and the identification of activity deviations to aid with personalised delivery of care.", "authors": ["Alexander Capstick", "Tianyu Cui", "Yu Chen", "Payam Barnaghi"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-05-07", "url": "https://arxiv.org/abs/2405.04494", "pdf_url": "https://arxiv.org/pdf/2405.04494v2", "arxiv_id": "2405.04494", "doi": "10.48550/arXiv.2405.04494", "citation_count": 2, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1193}
{"id": "492f6f10a388fce29666278d48d511b87dbd6c7ad4c4ae4b875340280ea5e749", "sources": ["arxiv", "semantic_scholar"], "title": "Beyond Unimodal Learning: The Importance of Integrating Multiple Modalities for Lifelong Learning", "abstract": "While humans excel at continual learning (CL), deep neural networks (DNNs) exhibit catastrophic forgetting. A salient feature of the brain that allows effective CL is that it utilizes multiple modalities for learning and inference, which is underexplored in DNNs. Therefore, we study the role and interactions of multiple modalities in mitigating forgetting and introduce a benchmark for multimodal continual learning. Our findings demonstrate that leveraging multiple views and complementary information from multiple modalities enables the model to learn more accurate and robust representations. This makes the model less vulnerable to modality-specific regularities and considerably mitigates forgetting. Furthermore, we observe that individual modalities exhibit varying degrees of robustness to distribution shift. Finally, we propose a method for integrating and aligning the information from different modalities by utilizing the relational structural similarities between the data points in each modality. Our method sets a strong baseline that enables both single- and multimodal inference. Our study provides a promising case for further exploring the role of multiple modalities in enabling CL and provides a standard benchmark for future research.", "authors": ["Fahad Sarfraz", "Bahram Zonooz", "Elahe Arani"], "categories": ["cs.LG", "cs.AI", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2024-05-04", "url": "https://arxiv.org/abs/2405.02766", "pdf_url": "https://arxiv.org/pdf/2405.02766v1", "arxiv_id": "2405.02766", "doi": "10.48550/arXiv.2405.02766", "citation_count": 5, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.1945}
{"id": "d173ab9b7187ad055440516f39d2dc3929e80394bdf778cac19acf16e77f5ab1", "sources": ["arxiv", "semantic_scholar"], "title": "IMEX-Reg: Implicit-Explicit Regularization in the Function Space for Continual Learning", "abstract": "Continual learning (CL) remains one of the long-standing challenges for deep neural networks due to catastrophic forgetting of previously acquired knowledge. Although rehearsal-based approaches have been fairly successful in mitigating catastrophic forgetting, they suffer from overfitting on buffered samples and prior information loss, hindering generalization under low-buffer regimes. Inspired by how humans learn using strong inductive biases, we propose IMEX-Reg to improve the generalization performance of experience rehearsal in CL under low buffer regimes. Specifically, we employ a two-pronged implicit-explicit regularization approach using contrastive representation learning (CRL) and consistency regularization. To further leverage the global relationship between representations learned using CRL, we propose a regularization strategy to guide the classifier toward the activation correlations in the unit hypersphere of the CRL. Our results show that IMEX-Reg significantly improves generalization performance and outperforms rehearsal-based approaches in several CL scenarios. It is also robust to natural and adversarial corruptions with less task-recency bias. Additionally, we provide theoretical insights to support our design decisions further.", "authors": ["Prashant Bhat", "Bharath Renjith", "Elahe Arani", "Bahram Zonooz"], "categories": ["cs.LG", "cs.AI", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2024-04-28", "url": "https://arxiv.org/abs/2404.18161", "pdf_url": "https://arxiv.org/pdf/2404.18161v1", "arxiv_id": "2404.18161", "doi": "10.48550/arXiv.2404.18161", "citation_count": 7, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2258}
{"id": "c1b7d767df551adf73456324b986eedfc2b399c09eef3b44e8d928a204c37cbe", "sources": ["arxiv", "semantic_scholar"], "title": "Addressing Loss of Plasticity and Catastrophic Forgetting in Continual Learning", "abstract": "Deep representation learning methods struggle with continual learning, suffering from both catastrophic forgetting of useful units and loss of plasticity, often due to rigid and unuseful units. While many methods address these two issues separately, only a few currently deal with both simultaneously. In this paper, we introduce Utility-based Perturbed Gradient Descent (UPGD) as a novel approach for the continual learning of representations. UPGD combines gradient updates with perturbations, where it applies smaller modifications to more useful units, protecting them from forgetting, and larger modifications to less useful units, rejuvenating their plasticity. We use a challenging streaming learning setup where continual learning problems have hundreds of non-stationarities and unknown task boundaries. We show that many existing methods suffer from at least one of the issues, predominantly manifested by their decreasing accuracy over tasks. On the other hand, UPGD continues to improve performance and surpasses or is competitive with all methods in all problems. Finally, in extended reinforcement learning experiments with PPO, we show that while Adam exhibits a performance drop after initial learning, UPGD avoids it by addressing both continual learning issues.", "authors": ["Mohamed Elsayed", "A. Rupam Mahmood"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-03-31", "url": "https://arxiv.org/abs/2404.00781", "pdf_url": "https://arxiv.org/pdf/2404.00781v2", "arxiv_id": "2404.00781", "doi": "10.48550/arXiv.2404.00781", "citation_count": 58, "influential_citation_count": 9, "has_code": true, "code_url": "https://github.com/mohmdelsayed/upgd", "venue": "International Conference on Learning Representations", "quality_score": 0.5}
{"id": "d90300b6c6235a28b89df1a083bee7f80844b5744fa8ec153ca2fe290403b4dd", "sources": ["arxiv", "semantic_scholar"], "title": "Function-space Parameterization of Neural Networks for Sequential Learning", "abstract": "Sequential learning paradigms pose challenges for gradient-based deep learning due to difficulties incorporating new data and retaining prior knowledge. While Gaussian processes elegantly tackle these problems, they struggle with scalability and handling rich inputs, such as images. To address these issues, we introduce a technique that converts neural networks from weight space to function space, through a dual parameterization. Our parameterization offers: (i) a way to scale function-space methods to large data sets via sparsification, (ii) retention of prior knowledge when access to past data is limited, and (iii) a mechanism to incorporate new data without retraining. Our experiments demonstrate that we can retain knowledge in continual learning and incorporate new data efficiently. We further show its strengths in uncertainty quantification and guiding exploration in model-based RL. Further information and code is available on the project website.", "authors": ["Aidan Scannell", "Riccardo Mereu", "Paul Chang", "Ella Tamir", "Joni Pajarinen", "Arno Solin"], "categories": ["stat.ML", "cs.LG"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2024-03-16", "url": "https://arxiv.org/abs/2403.10929", "pdf_url": "https://arxiv.org/pdf/2403.10929v1", "arxiv_id": "2403.10929", "doi": "10.48550/arXiv.2403.10929", "citation_count": 8, "influential_citation_count": 0, "has_code": true, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.2386}
{"id": "8bd80c25b16808e45475179bdc6664c64a01ca0143f8ecfc8d36953afff09e76", "sources": ["arxiv", "semantic_scholar"], "title": "Continual Learning and Catastrophic Forgetting", "abstract": "This book chapter delves into the dynamics of continual learning, which is the process of incrementally learning from a non-stationary stream of data. Although continual learning is a natural skill for the human brain, it is very challenging for artificial neural networks. An important reason is that, when learning something new, these networks tend to quickly and drastically forget what they had learned before, a phenomenon known as catastrophic forgetting. Especially in the last decade, continual learning has become an extensively studied topic in deep learning. This book chapter reviews the insights that this field has generated.", "authors": ["Gido M. van de Ven", "Nicholas Soures", "Dhireesha Kudithipudi"], "categories": ["cs.LG", "cs.AI", "cs.CV", "q-bio.NC", "stat.ML"], "fields_of_study": ["Computer Science", "Biology", "Mathematics"], "published_date": "2024-03-08", "url": "https://arxiv.org/abs/2403.05175", "pdf_url": "https://arxiv.org/pdf/2403.05175v1", "arxiv_id": "2403.05175", "doi": "10.1016/B978-0-443-15754-7.00073-0", "citation_count": 130, "influential_citation_count": 9, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.5293}
{"id": "42f7136df5079c8ccebed9643d8cb9c238544f97ca34b5a77f96da41ec39ba86", "sources": ["arxiv", "semantic_scholar"], "title": "Gradient Correlation Subspace Learning against Catastrophic Forgetting", "abstract": "Efficient continual learning techniques have been a topic of significant research over the last few years. A fundamental problem with such learning is severe degradation of performance on previously learned tasks, known also as catastrophic forgetting. This paper introduces a novel method to reduce catastrophic forgetting in the context of incremental class learning called Gradient Correlation Subspace Learning (GCSL). The method detects a subspace of the weights that is least affected by previous tasks and projects the weights to train for the new task into said subspace. The method can be applied to one or more layers of a given network architectures and the size of the subspace used can be altered from layer to layer and task to task. Code will be available at \\href{https://github.com/vgthengane/GCSL}{https://github.com/vgthengane/GCSL}", "authors": ["Tammuz Dubnov", "Vishal Thengane"], "categories": ["cs.LG", "cs.AI", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2024-03-04", "url": "https://arxiv.org/abs/2403.02334", "pdf_url": "https://arxiv.org/pdf/2403.02334v1", "arxiv_id": "2403.02334", "doi": "10.48550/arXiv.2403.02334", "citation_count": 0, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/vgthengane/GCSL", "venue": "arXiv.org", "quality_score": 0.0}
{"id": "38586accf225602256ef76e8bac601f05814f655007136f45774d07bb9e66f88", "sources": ["arxiv", "semantic_scholar"], "title": "Decoupled Subgraph Federated Learning", "abstract": "We address the challenge of federated learning on graph-structured data distributed across multiple clients. Specifically, we focus on the prevalent scenario of interconnected subgraphs, where interconnections between different clients play a critical role. We present a novel framework for this scenario, named FedStruct, that harnesses deep structural dependencies. To uphold privacy, unlike existing methods, FedStruct eliminates the necessity of sharing or generating sensitive node features or embeddings among clients. Instead, it leverages explicit global graph structure information to capture inter-node dependencies. We validate the effectiveness of FedStruct through experimental results conducted on six datasets for semi-supervised node classification, showcasing performance close to the centralized approach across various scenarios, including different data partitioning methods, varying levels of label availability, and number of clients.", "authors": ["Javad Aliakbari", "Johan Östman", "Alexandre Graell i Amat"], "categories": ["cs.LG", "cs.IT"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2024-02-29", "url": "https://arxiv.org/abs/2402.19163", "pdf_url": "https://arxiv.org/pdf/2402.19163v3", "arxiv_id": "2402.19163", "doi": null, "citation_count": 3, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.1505}
{"id": "5994a936a799674840e9314965a053ecb2b28cd9a584e5af0fec51c0c19805be", "sources": ["arxiv", "semantic_scholar"], "title": "Hebbian Learning based Orthogonal Projection for Continual Learning of Spiking Neural Networks", "abstract": "Neuromorphic computing with spiking neural networks is promising for energy-efficient artificial intelligence (AI) applications. However, different from humans who continually learn different tasks in a lifetime, neural network models suffer from catastrophic forgetting. How could neuronal operations solve this problem is an important question for AI and neuroscience. Many previous studies draw inspiration from observed neuroscience phenomena and propose episodic replay or synaptic metaplasticity, but they are not guaranteed to explicitly preserve knowledge for neuron populations. Other works focus on machine learning methods with more mathematical grounding, e.g., orthogonal projection on high dimensional spaces, but there is no neural correspondence for neuromorphic computing. In this work, we develop a new method with neuronal operations based on lateral connections and Hebbian learning, which can protect knowledge by projecting activity traces of neurons into an orthogonal subspace so that synaptic weight update will not interfere with old tasks. We show that Hebbian and anti-Hebbian learning on recurrent lateral connections can effectively extract the principal subspace of neural activities and enable orthogonal projection. This provides new insights into how neural circuits and Hebbian learning can help continual learning, and also how the concept of orthogonal projection can be realized in neuronal systems. Our method is also flexible to utilize arbitrary training methods based on presynaptic activities/traces. Experiments show that our method consistently solves forgetting for spiking neural networks with nearly zero forgetting under various supervised training methods with different error propagation approaches, and outperforms previous approaches under various settings. Our method can pave a solid path for building continual neuromorphic computing systems.", "authors": ["Mingqing Xiao", "Qingyan Meng", "Zongpeng Zhang", "Di He", "Zhouchen Lin"], "categories": ["cs.NE", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-02-19", "url": "https://arxiv.org/abs/2402.11984", "pdf_url": "https://arxiv.org/pdf/2402.11984v1", "arxiv_id": "2402.11984", "doi": "10.48550/arXiv.2402.11984", "citation_count": 18, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.3197}
{"id": "9e44ff5f0b78012c4564cadbadf53ae584500547d7cb9b0e45e2146b0325f903", "sources": ["arxiv", "semantic_scholar"], "title": "ResQuNNs: Towards Enabling Deep Learning in Quantum Convolution Neural Networks", "abstract": "In this paper, we present a novel framework for enhancing the performance of Quanvolutional Neural Networks (QuNNs) by introducing trainable quanvolutional layers and addressing the critical challenges associated with them. Traditional quanvolutional layers, although beneficial for feature extraction, have largely been static, offering limited adaptability. Unlike state-of-the-art, our research overcomes this limitation by enabling training within these layers, significantly increasing the flexibility and potential of QuNNs. However, the introduction of multiple trainable quanvolutional layers induces complexities in gradient-based optimization, primarily due to the difficulty in accessing gradients across these layers. To resolve this, we propose a novel architecture, Residual Quanvolutional Neural Networks (ResQuNNs), leveraging the concept of residual learning, which facilitates the flow of gradients by adding skip connections between layers. By inserting residual blocks between quanvolutional layers, we ensure enhanced gradient access throughout the network, leading to improved training performance. Moreover, we provide empirical evidence on the strategic placement of these residual blocks within QuNNs. Through extensive experimentation, we identify an efficient configuration of residual blocks, which enables gradients across all the layers in the network that eventually results in efficient training. Our findings suggest that the precise location of residual blocks plays a crucial role in maximizing the performance gains in QuNNs. Our results mark a substantial step forward in the evolution of quantum deep learning, offering new avenues for both theoretical development and practical quantum computing applications.", "authors": ["Muhammad Kashif", "Muhammad Shafique"], "categories": ["cs.LG", "cs.AI", "quant-ph"], "fields_of_study": ["Medicine", "Computer Science", "Physics"], "published_date": "2024-02-14", "url": "https://arxiv.org/abs/2402.09146", "pdf_url": "https://arxiv.org/pdf/2402.09146v6", "arxiv_id": "2402.09146", "doi": "10.1038/s41598-025-06035-4", "citation_count": 9, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Scientific Reports", "quality_score": 0.25}
{"id": "a5abd3a9d385cd33f0ba7447575f418e0076ee38c539906b71974ba8d3ae6c36", "sources": ["arxiv", "semantic_scholar"], "title": "Unveiling Group-Specific Distributed Concept Drift: A Fairness Imperative in Federated Learning", "abstract": "In the evolving field of machine learning, ensuring group fairness has become a critical concern, prompting the development of algorithms designed to mitigate bias in decision-making processes. Group fairness refers to the principle that a model's decisions should be equitable across different groups defined by sensitive attributes such as gender or race, ensuring that individuals from privileged groups and unprivileged groups are treated fairly and receive similar outcomes. However, achieving fairness in the presence of group-specific concept drift remains an unexplored frontier, and our research represents pioneering efforts in this regard. Group-specific concept drift refers to situations where one group experiences concept drift over time while another does not, leading to a decrease in fairness even if accuracy remains fairly stable. Within the framework of Federated Learning, where clients collaboratively train models, its distributed nature further amplifies these challenges since each client can experience group-specific concept drift independently while still sharing the same underlying concept, creating a complex and dynamic environment for maintaining fairness. The most significant contribution of our research is the formalization and introduction of the problem of group-specific concept drift and its distributed counterpart, shedding light on its critical importance in the field of fairness. Additionally, leveraging insights from prior research, we adapt an existing distributed concept drift adaptation algorithm to tackle group-specific distributed concept drift which uses a multi-model approach, a local group-specific drift detection mechanism, and continuous clustering of models over time. The findings from our experiments highlight the importance of addressing group-specific concept drift and its distributed counterpart to advance fairness in machine learning.", "authors": ["Teresa Salazar", "João Gama", "Helder Araújo", "Pedro Henriques Abreu"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science", "Medicine"], "published_date": "2024-02-12", "url": "https://arxiv.org/abs/2402.07586", "pdf_url": "https://arxiv.org/pdf/2402.07586v4", "arxiv_id": "2402.07586", "doi": "10.1109/TNNLS.2025.3601834", "citation_count": 10, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.2603}
{"id": "a4f041a6d7d0f41585ffd5efbf6fcfb2ed0439e3b690cc62d3f46fc04292142d", "sources": ["arxiv", "semantic_scholar"], "title": "Score-based Causal Representation Learning: Linear and General Transformations", "abstract": "This paper addresses intervention-based causal representation learning (CRL) under a general nonparametric latent causal model and an unknown transformation that maps the latent variables to the observed variables. Linear and general transformations are investigated. The paper addresses both the identifiability and achievability aspects. Identifiability refers to determining algorithm-agnostic conditions that ensure the recovery of the true latent causal variables and the underlying latent causal graph. Achievability refers to the algorithmic aspects and addresses designing algorithms that achieve identifiability guarantees. By drawing novel connections between score functions (i.e., the gradients of the logarithm of density functions) and CRL, this paper designs a score-based class of algorithms that ensures both identifiability and achievability. First, the paper focuses on linear transformations and shows that one stochastic hard intervention per node suffices to guarantee identifiability. It also provides partial identifiability guarantees for soft interventions, including identifiability up to mixing with parents for general causal models and perfect recovery of the latent graph for sufficiently nonlinear causal models. Secondly, it focuses on general transformations and demonstrates that two stochastic hard interventions per node are sufficient for identifiability. This is achieved by defining a differentiable loss function whose global optima ensure identifiability for general CRL. Notably, one does not need to know which pair of interventional environments has the same node intervened. Finally, the theoretical results are empirically validated via experiments on structured synthetic data and image data.", "authors": ["Burak Varıcı", "Emre Acartürk", "Karthikeyan Shanmugam", "Abhishek Kumar", "Ali Tajer"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2024-02-01", "url": "https://arxiv.org/abs/2402.00849", "pdf_url": "https://arxiv.org/pdf/2402.00849v5", "arxiv_id": "2402.00849", "doi": "10.48550/arXiv.2402.00849", "citation_count": 28, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3656}
{"id": "5dd41c1bd0bebc052d56b0e1315a0d2ab7d4a2ae268745adcb6490d5a0adbd65", "sources": ["arxiv", "semantic_scholar"], "title": "The Joint Effect of Task Similarity and Overparameterization on Catastrophic Forgetting -- An Analytical Model", "abstract": "In continual learning, catastrophic forgetting is affected by multiple aspects of the tasks. Previous works have analyzed separately how forgetting is affected by either task similarity or overparameterization. In contrast, our paper examines how task similarity and overparameterization jointly affect forgetting in an analyzable model. Specifically, we focus on two-task continual linear regression, where the second task is a random orthogonal transformation of an arbitrary first task (an abstraction of random permutation tasks). We derive an exact analytical expression for the expected forgetting - and uncover a nuanced pattern. In highly overparameterized models, intermediate task similarity causes the most forgetting. However, near the interpolation threshold, forgetting decreases monotonically with the expected task similarity. We validate our findings with linear regression on synthetic data, and with neural networks on established permutation task benchmarks.", "authors": ["Daniel Goldfarb", "Itay Evron", "Nir Weinberger", "Daniel Soudry", "Paul Hand"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-01-23", "url": "https://arxiv.org/abs/2401.12617", "pdf_url": "https://arxiv.org/pdf/2401.12617v2", "arxiv_id": "2401.12617", "doi": "10.48550/arXiv.2401.12617", "citation_count": 26, "influential_citation_count": 3, "has_code": false, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.3578}
{"id": "b65654f7f82efdd8c24dcbfbfb1b5adfdf3a4940ce71e5e5625fcd9cde1af51f", "sources": ["arxiv", "semantic_scholar"], "title": "t-DGR: A Trajectory-Based Deep Generative Replay Method for Continual Learning in Decision Making", "abstract": "Deep generative replay has emerged as a promising approach for continual learning in decision-making tasks. This approach addresses the problem of catastrophic forgetting by leveraging the generation of trajectories from previously encountered tasks to augment the current dataset. However, existing deep generative replay methods for continual learning rely on autoregressive models, which suffer from compounding errors in the generated trajectories. In this paper, we propose a simple, scalable, and non-autoregressive method for continual learning in decision-making tasks using a generative model that generates task samples conditioned on the trajectory timestep. We evaluate our method on Continual World benchmarks and find that our approach achieves state-of-the-art performance on the average success rate metric among continual learning methods. Code is available at https://github.com/WilliamYue37/t-DGR.", "authors": ["William Yue", "Bo Liu", "Peter Stone"], "categories": ["cs.LG", "cs.AI", "cs.NE"], "fields_of_study": ["Computer Science"], "published_date": "2024-01-04", "url": "https://arxiv.org/abs/2401.02576", "pdf_url": "https://arxiv.org/pdf/2401.02576v2", "arxiv_id": "2401.02576", "doi": "10.48550/arXiv.2401.02576", "citation_count": 7, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/WilliamYue37/t-DGR", "venue": null, "quality_score": 0.2258}
{"id": "28a60911d1f3436270e21aad96f61e28dfa15f73474de9b0deb2795988644a8f", "sources": ["arxiv", "semantic_scholar"], "title": "Continual Learning via Sequential Function-Space Variational Inference", "abstract": "Sequential Bayesian inference over predictive functions is a natural framework for continual learning from streams of data. However, applying it to neural networks has proved challenging in practice. Addressing the drawbacks of existing techniques, we propose an optimization objective derived by formulating continual learning as sequential function-space variational inference. In contrast to existing methods that regularize neural network parameters directly, this objective allows parameters to vary widely during training, enabling better adaptation to new tasks. Compared to objectives that directly regularize neural network predictions, the proposed objective allows for more flexible variational distributions and more effective regularization. We demonstrate that, across a range of task sequences, neural networks trained via sequential function-space variational inference achieve better predictive accuracy than networks trained with related methods while depending less on maintaining a set of representative points from previous tasks.", "authors": ["Tim G. J. Rudner", "Freddie Bickford Smith", "Qixuan Feng", "Yee Whye Teh", "Yarin Gal"], "categories": ["stat.ML", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2023-12-28", "url": "https://arxiv.org/abs/2312.17210", "pdf_url": "https://arxiv.org/pdf/2312.17210v1", "arxiv_id": "2312.17210", "doi": "10.48550/arXiv.2312.17210", "citation_count": 58, "influential_citation_count": 6, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.4427}
{"id": "c1592ee131eecabb9b407d6b48c7a2ec5f52f8a15120331f55cf4e64b9b78a05", "sources": ["arxiv", "semantic_scholar"], "title": "Infinite dSprites for Disentangled Continual Learning: Separating Memory Edits from Generalization", "abstract": "The ability of machine learning systems to learn continually is hindered by catastrophic forgetting, the tendency of neural networks to overwrite previously acquired knowledge when learning a new task. Existing methods mitigate this problem through regularization, parameter isolation, or rehearsal, but they are typically evaluated on benchmarks comprising only a handful of tasks. In contrast, humans are able to learn over long time horizons in dynamic, open-world environments, effortlessly memorizing unfamiliar objects and reliably recognizing them under various transformations. To make progress towards closing this gap, we introduce Infinite dSprites, a parsimonious tool for creating continual classification and disentanglement benchmarks of arbitrary length and with full control over generative factors. We show that over a sufficiently long time horizon, the performance of all major types of continual learning methods deteriorates on this simple benchmark. This result highlights an important and previously overlooked aspect of continual learning: given a finite modelling capacity and an arbitrarily long learning horizon, efficient learning requires memorizing class-specific information and accumulating knowledge about general mechanisms. In a simple setting with direct supervision on the generative factors, we show how learning class-agnostic transformations offers a way to circumvent catastrophic forgetting and improve classification accuracy over time. Our approach sets the stage for continual learning over hundreds of tasks with explicit control over memorization and forgetting, emphasizing open-set classification and one-shot generalization.", "authors": ["Sebastian Dziadzio", "Çağatay Yıldız", "Gido M. van de Ven", "Tomasz Trzciński", "Tinne Tuytelaars", "Matthias Bethge"], "categories": ["cs.LG", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2023-12-27", "url": "https://arxiv.org/abs/2312.16731", "pdf_url": "https://arxiv.org/pdf/2312.16731v3", "arxiv_id": "2312.16731", "doi": null, "citation_count": 2, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Proceedings of The 3rd Conference on Lifelong Learning Agents, PMLR 274: 498-513, 2025", "quality_score": 0.1193}
{"id": "9b9f4349e3af0697b69494dc638d61b8f54c384f5170d1546f413679f6711b3a", "sources": ["arxiv", "semantic_scholar"], "title": "Temporal Supervised Contrastive Learning for Modeling Patient Risk Progression", "abstract": "We consider the problem of predicting how the likelihood of an outcome of interest for a patient changes over time as we observe more of the patient data. To solve this problem, we propose a supervised contrastive learning framework that learns an embedding representation for each time step of a patient time series. Our framework learns the embedding space to have the following properties: (1) nearby points in the embedding space have similar predicted class probabilities, (2) adjacent time steps of the same time series map to nearby points in the embedding space, and (3) time steps with very different raw feature vectors map to far apart regions of the embedding space. To achieve property (3), we employ a nearest neighbor pairing mechanism in the raw feature space. This mechanism also serves as an alternative to data augmentation, a key ingredient of contrastive learning, which lacks a standard procedure that is adequately realistic for clinical tabular data, to our knowledge. We demonstrate that our approach outperforms state-of-the-art baselines in predicting mortality of septic patients (MIMIC-III dataset) and tracking progression of cognitive impairment (ADNI dataset). Our method also consistently recovers the correct synthetic dataset embedding structure across experiments, a feat not achieved by baselines. Our ablation experiments show the pivotal role of our nearest neighbor pairing.", "authors": ["Shahriar Noroozizadeh", "Jeremy C. Weiss", "George H. Chen"], "categories": ["cs.LG", "cs.AI", "stat.ML"], "fields_of_study": ["Medicine", "Computer Science", "Mathematics"], "published_date": "2023-12-10", "url": "https://arxiv.org/abs/2312.05933", "pdf_url": "https://arxiv.org/pdf/2312.05933v1", "arxiv_id": "2312.05933", "doi": "10.48550/arXiv.2312.05933", "citation_count": 10, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "In Machine Learning for Health (ML4H), pages 403-427. PMLR, 2023", "quality_score": 0.2603}
{"id": "ccbbea68f1436ac18c9bc7140dcedccf9c92c61be5f75c1795c78da0752ae7bf", "sources": ["arxiv", "semantic_scholar"], "title": "Metalearning Continual Learning Algorithms", "abstract": "General-purpose learning systems should improve themselves in open-ended fashion in ever-changing environments. Conventional learning algorithms for neural networks, however, suffer from catastrophic forgetting (CF), i.e., previously acquired skills are forgotten when a new task is learned. Instead of hand-crafting new algorithms for avoiding CF, we propose Automated Continual Learning (ACL) to train self-referential neural networks to metalearn their own in-context continual (meta)learning algorithms. ACL encodes continual learning (CL) desiderata -- good performance on both old and new tasks -- into its metalearning objectives. Our experiments demonstrate that ACL effectively resolves \"in-context catastrophic forgetting,\" a problem that naive in-context learning algorithms suffer from; ACL-learned algorithms outperform both hand-crafted learning algorithms and popular meta-continual learning methods on the Split-MNIST benchmark in the replay-free setting, and enables continual learning of diverse tasks consisting of multiple standard image classification datasets. We also discuss the current limitations of in-context CL by comparing ACL with state-of-the-art CL methods that leverage pre-trained models. Overall, we bring several novel perspectives into the long-standing problem of CL.", "authors": ["Kazuki Irie", "Róbert Csordás", "Jürgen Schmidhuber"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2023-12-01", "url": "https://arxiv.org/abs/2312.00276", "pdf_url": "https://arxiv.org/pdf/2312.00276v3", "arxiv_id": "2312.00276", "doi": null, "citation_count": 8, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2386}
{"id": "fd53cd33060cc2bb28b59d79c888f0e8b3496e8e60a23abeeed5141889259f6a", "sources": ["arxiv", "semantic_scholar"], "title": "Continual Learning of Diffusion Models with Generative Distillation", "abstract": "Diffusion models are powerful generative models that achieve state-of-the-art performance in image synthesis. However, training them demands substantial amounts of data and computational resources. Continual learning would allow for incrementally learning new tasks and accumulating knowledge, thus enabling the reuse of trained models for further learning. One potentially suitable continual learning approach is generative replay, where a copy of a generative model trained on previous tasks produces synthetic data that are interleaved with data from the current task. However, standard generative replay applied to diffusion models results in a catastrophic loss in denoising capabilities. In this paper, we propose generative distillation, an approach that distils the entire reverse process of a diffusion model. We demonstrate that our approach substantially improves the continual learning performance of generative replay with only a modest increase in the computational costs.", "authors": ["Sergi Masip", "Pau Rodriguez", "Tinne Tuytelaars", "Gido M. van de Ven"], "categories": ["cs.LG", "cs.AI", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2023-11-23", "url": "https://arxiv.org/abs/2311.14028", "pdf_url": "https://arxiv.org/pdf/2311.14028v2", "arxiv_id": "2311.14028", "doi": "10.48550/arXiv.2311.14028", "citation_count": 24, "influential_citation_count": 3, "has_code": false, "code_url": null, "venue": "Proceedings of The 3rd Conference on Lifelong Learning Agents, PMLR 274: 431-456, 2025", "quality_score": 0.3495}
{"id": "d156bef5750ce166e7fa237ce13e061d32879f874bb171b343a645a586f9fee9", "sources": ["arxiv", "semantic_scholar"], "title": "Replay-enhanced Continual Reinforcement Learning", "abstract": "Replaying past experiences has proven to be a highly effective approach for averting catastrophic forgetting in supervised continual learning. However, some crucial factors are still largely ignored, making it vulnerable to serious failure, when used as a solution to forgetting in continual reinforcement learning, even in the context of perfect memory where all data of previous tasks are accessible in the current task. On the one hand, since most reinforcement learning algorithms are not invariant to the reward scale, the previously well-learned tasks (with high rewards) may appear to be more salient to the current learning process than the current task (with small initial rewards). This causes the agent to concentrate on those salient tasks at the expense of generality on the current task. On the other hand, offline learning on replayed tasks while learning a new task may induce a distributional shift between the dataset and the learned policy on old tasks, resulting in forgetting. In this paper, we introduce RECALL, a replay-enhanced method that greatly improves the plasticity of existing replay-based methods on new tasks while effectively avoiding the recurrence of catastrophic forgetting in continual reinforcement learning. RECALL leverages adaptive normalization on approximate targets and policy distillation on old tasks to enhance generality and stability, respectively. Extensive experiments on the Continual World benchmark show that RECALL performs significantly better than purely perfect memory replay, and achieves comparable or better overall performance against state-of-the-art continual learning methods.", "authors": ["Tiantian Zhang", "Kevin Zehua Shen", "Zichuan Lin", "Bo Yuan", "Xueqian Wang", "Xiu Li", "Deheng Ye"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2023-11-20", "url": "https://arxiv.org/abs/2311.11557", "pdf_url": "https://arxiv.org/pdf/2311.11557v1", "arxiv_id": "2311.11557", "doi": "10.48550/arXiv.2311.11557", "citation_count": 12, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2785}
{"id": "a5210e14dcb072cd84636f699a30a4de143bc403235df6357c6851cbe00fbf0b", "sources": ["arxiv", "semantic_scholar"], "title": "Online Continual Learning via Logit Adjusted Softmax", "abstract": "Online continual learning is a challenging problem where models must learn from a non-stationary data stream while avoiding catastrophic forgetting. Inter-class imbalance during training has been identified as a major cause of forgetting, leading to model prediction bias towards recently learned classes. In this paper, we theoretically analyze that inter-class imbalance is entirely attributed to imbalanced class-priors, and the function learned from intra-class intrinsic distributions is the Bayes-optimal classifier. To that end, we present that a simple adjustment of model logits during training can effectively resist prior class bias and pursue the corresponding Bayes-optimum. Our proposed method, Logit Adjusted Softmax, can mitigate the impact of inter-class imbalance not only in class-incremental but also in realistic general setups, with little additional computational cost. We evaluate our approach on various benchmarks and demonstrate significant performance improvements compared to prior arts. For example, our approach improves the best baseline by 4.6% on CIFAR10.", "authors": ["Zhehao Huang", "Tao Li", "Chenhe Yuan", "Yingwen Wu", "Xiaolin Huang"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2023-11-11", "url": "https://arxiv.org/abs/2311.06460", "pdf_url": "https://arxiv.org/pdf/2311.06460v2", "arxiv_id": "2311.06460", "doi": "10.48550/arXiv.2311.06460", "citation_count": 6, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2113}
{"id": "392546d3b734d49fff146398a734feb2214e5d8792251974568aea14904c6f6e", "sources": ["arxiv", "semantic_scholar"], "title": "Implicit biases in multitask and continual learning from a backward error analysis perspective", "abstract": "Using backward error analysis, we compute implicit training biases in multitask and continual learning settings for neural networks trained with stochastic gradient descent. In particular, we derive modified losses that are implicitly minimized during training. They have three terms: the original loss, accounting for convergence, an implicit flatness regularization term proportional to the learning rate, and a last term, the conflict term, which can theoretically be detrimental to both convergence and implicit regularization. In multitask, the conflict term is a well-known quantity, measuring the gradient alignment between the tasks, while in continual learning the conflict term is a new quantity in deep learning optimization, although a basic tool in differential geometry: The Lie bracket between the task gradients.", "authors": ["Benoit Dherin"], "categories": ["stat.ML", "cs.AI", "cs.LG"], "fields_of_study": ["Mathematics", "Computer Science"], "published_date": "2023-11-01", "url": "https://arxiv.org/abs/2311.00235", "pdf_url": "https://arxiv.org/pdf/2311.00235v1", "arxiv_id": "2311.00235", "doi": "10.48550/arXiv.2311.00235", "citation_count": 3, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1505}
{"id": "3f33753322947b21dc489a6d8bb4b4ffbc6c7f951131aa3b645d569a94e8fed2", "sources": ["arxiv", "semantic_scholar"], "title": "GOPlan: Goal-conditioned Offline Reinforcement Learning by Planning with Learned Models", "abstract": "Offline Goal-Conditioned RL (GCRL) offers a feasible paradigm for learning general-purpose policies from diverse and multi-task offline datasets. Despite notable recent progress, the predominant offline GCRL methods, mainly model-free, face constraints in handling limited data and generalizing to unseen goals. In this work, we propose Goal-conditioned Offline Planning (GOPlan), a novel model-based framework that contains two key phases: (1) pretraining a prior policy capable of capturing multi-modal action distribution within the multi-goal dataset; (2) employing the reanalysis method with planning to generate imagined trajectories for funetuning policies. Specifically, we base the prior policy on an advantage-weighted conditioned generative adversarial network, which facilitates distinct mode separation, mitigating the pitfalls of out-of-distribution (OOD) actions. For further policy optimization, the reanalysis method generates high-quality imaginary data by planning with learned models for both intra-trajectory and inter-trajectory goals. With thorough experimental evaluations, we demonstrate that GOPlan achieves state-of-the-art performance on various offline multi-goal navigation and manipulation tasks. Moreover, our results highlight the superior ability of GOPlan to handle small data budgets and generalize to OOD goals.", "authors": ["Mianchu Wang", "Rui Yang", "Xi Chen", "Hao Sun", "Meng Fang", "Giovanni Montana"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2023-10-30", "url": "https://arxiv.org/abs/2310.20025", "pdf_url": "https://arxiv.org/pdf/2310.20025v3", "arxiv_id": "2310.20025", "doi": "10.48550/arXiv.2310.20025", "citation_count": 20, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Transactions on Machine Learning Research (05/2024)", "quality_score": 0.3306}
{"id": "4fcbca39f186b271c49f50655914298de4c7e981fea79dd4f9a842df278640cc", "sources": ["arxiv", "semantic_scholar"], "title": "Delayed Memory Unit: Modelling Temporal Dependency Through Delay Gate", "abstract": "Recurrent Neural Networks (RNNs) are widely recognized for their proficiency in modeling temporal dependencies, making them highly prevalent in sequential data processing applications. Nevertheless, vanilla RNNs are confronted with the well-known issue of gradient vanishing and exploding, posing a significant challenge for learning and establishing long-range dependencies. Additionally, gated RNNs tend to be over-parameterized, resulting in poor computational efficiency and network generalization. To address these challenges, this paper proposes a novel Delayed Memory Unit (DMU). The DMU incorporates a delay line structure along with delay gates into vanilla RNN, thereby enhancing temporal interaction and facilitating temporal credit assignment. Specifically, the DMU is designed to directly distribute the input information to the optimal time instant in the future, rather than aggregating and redistributing it over time through intricate network dynamics. Our proposed DMU demonstrates superior temporal modeling capabilities across a broad range of sequential modeling tasks, utilizing considerably fewer parameters than other state-of-the-art gated RNN models in applications such as speech recognition, radar gesture recognition, ECG waveform segmentation, and permuted sequential image classification.", "authors": ["Pengfei Sun", "Jibin Wu", "Malu Zhang", "Paul Devos", "Dick Botteldooren"], "categories": ["cs.NE", "cs.LG", "eess.AS", "eess.SP"], "fields_of_study": ["Computer Science", "Medicine", "Engineering"], "published_date": "2023-10-23", "url": "https://arxiv.org/abs/2310.14982", "pdf_url": "https://arxiv.org/pdf/2310.14982v2", "arxiv_id": "2310.14982", "doi": "10.1109/TNNLS.2024.3490833", "citation_count": 20, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.3306}
{"id": "c1f777ba5de5a88a448fb0dcabb85d37cc084cb6a88ea59b02b012060387597d", "sources": ["arxiv", "semantic_scholar"], "title": "Dual Cognitive Architecture: Incorporating Biases and Multi-Memory Systems for Lifelong Learning", "abstract": "Artificial neural networks (ANNs) exhibit a narrow scope of expertise on stationary independent data. However, the data in the real world is continuous and dynamic, and ANNs must adapt to novel scenarios while also retaining the learned knowledge to become lifelong learners. The ability of humans to excel at these tasks can be attributed to multiple factors ranging from cognitive computational structures, cognitive biases, and the multi-memory systems in the brain. We incorporate key concepts from each of these to design a novel framework, Dual Cognitive Architecture (DUCA), which includes multiple sub-systems, implicit and explicit knowledge representation dichotomy, inductive bias, and a multi-memory system. The inductive bias learner within DUCA is instrumental in encoding shape information, effectively countering the tendency of ANNs to learn local textures. Simultaneously, the inclusion of a semantic memory submodule facilitates the gradual consolidation of knowledge, replicating the dynamics observed in fast and slow learning systems, reminiscent of the principles underpinning the complementary learning system in human cognition. DUCA shows improvement across different settings and datasets, and it also exhibits reduced task recency bias, without the need for extra information. To further test the versatility of lifelong learning methods on a challenging distribution shift, we introduce a novel domain-incremental dataset DN4IL. In addition to improving performance on existing benchmarks, DUCA also demonstrates superior performance on this complex dataset.", "authors": ["Shruthi Gowda", "Bahram Zonooz", "Elahe Arani"], "categories": ["cs.CV", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2023-10-17", "url": "https://arxiv.org/abs/2310.11341", "pdf_url": "https://arxiv.org/pdf/2310.11341v1", "arxiv_id": "2310.11341", "doi": "10.48550/arXiv.2310.11341", "citation_count": 6, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2113}
{"id": "b467b88190e60a886c0ecc165599dcb0f72c636d8c4f7f54895839a852e6fabd", "sources": ["arxiv", "semantic_scholar"], "title": "Investigating Continuous Learning in Spiking Neural Networks", "abstract": "In this paper, the use of third-generation machine learning, also known as spiking neural network architecture, for continuous learning was investigated and compared to conventional models. The experimentation was divided into three separate phases. The first phase focused on training the conventional models via transfer learning. The second phase trains a Nengo model from their library. Lastly, each conventional model is converted into a spiking neural network and trained. Initial results from phase 1 are inline with known knowledge about continuous learning within current machine learning literature. All models were able to correctly identify the current classes, but they would immediately see a sharp performance drop in previous classes due to catastrophic forgetting. However, the SNN models were able to retain some information about previous classes. Although many of the previous classes were still identified as the current trained classes, the output probabilities showed a higher than normal value to the actual class. This indicates that the SNN models do have potential to overcome catastrophic forgetting but much work is still needed.", "authors": ["C. Tanner Fredieu"], "categories": ["cs.NE", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2023-10-09", "url": "https://arxiv.org/abs/2310.05343", "pdf_url": "https://arxiv.org/pdf/2310.05343v1", "arxiv_id": "2310.05343", "doi": "10.48550/arXiv.2310.05343", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.0}
{"id": "1f3d7158149dba1394e5d6128d34542c358ed33d35b02176658c56e0c28891eb", "sources": ["arxiv", "semantic_scholar"], "title": "Chunking: Continual Learning is not just about Distribution Shift", "abstract": "Work on continual learning (CL) has thus far largely focused on the problems arising from shifts in the data distribution. However, CL can be decomposed into two sub-problems: (a) shifts in the data distribution, and (b) dealing with the fact that the data is split into chunks and so only a part of the data is available to be trained on at any point in time. In this work, we look at the latter sub-problem, the chunking of data. We show that chunking is an important part of CL, accounting for around half of the performance drop from offline learning in our experiments. Furthermore, our results reveal that current CL algorithms do not address the chunking sub-problem, only performing as well as plain SGD training when there is no shift in the data distribution. Therefore, we show that chunking is both an important and currently unaddressed sub-problem and until it is addressed CL methods will be capped in performance. Additionally, we analyse why performance drops when learning occurs on identically distributed chunks of data, and find that forgetting, which is often seen to be a problem due to distribution shift, still arises and is a significant problem. We also show that performance on the chunking sub-problem can be increased and that this performance transfers to the full CL setting, where there is distribution shift. Hence, we argue that work on chunking can help advance CL in general.", "authors": ["Thomas L. Lee", "Amos Storkey"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2023-10-03", "url": "https://arxiv.org/abs/2310.02206", "pdf_url": "https://arxiv.org/pdf/2310.02206v2", "arxiv_id": "2310.02206", "doi": null, "citation_count": 2, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.1193}
{"id": "c0874b169f6cdd24a77db04ac1b85ee472981adae6b0ed8fde412c351b242df9", "sources": ["arxiv", "semantic_scholar"], "title": "Continual Evidential Deep Learning for Out-of-Distribution Detection", "abstract": "Uncertainty-based deep learning models have attracted a great deal of interest for their ability to provide accurate and reliable predictions. Evidential deep learning stands out achieving remarkable performance in detecting out-of-distribution (OOD) data with a single deterministic neural network. Motivated by this fact, in this paper we propose the integration of an evidential deep learning method into a continual learning framework in order to perform simultaneously incremental object classification and OOD detection. Moreover, we analyze the ability of vacuity and dissonance to differentiate between in-distribution data belonging to old classes and OOD data. The proposed method, called CEDL, is evaluated on CIFAR-100 considering two settings consisting of 5 and 10 tasks, respectively. From the obtained results, we could appreciate that the proposed method, in addition to provide comparable results in object classification with respect to the baseline, largely outperforms OOD detection compared to several posthoc methods on three evaluation metrics: AUROC, AUPR and FPR95.", "authors": ["Eduardo Aguilar", "Bogdan Raducanu", "Petia Radeva", "Joost Van de Weijer"], "categories": ["cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2023-09-06", "url": "https://arxiv.org/abs/2309.02995", "pdf_url": "https://arxiv.org/pdf/2309.02995v1", "arxiv_id": "2309.02995", "doi": "10.1109/ICCVW60793.2023.00369", "citation_count": 19, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.3253}
{"id": "3fdc9bf905df76e79b3d8a9fc38f8ecd230fbb215e5dfdb9d2501e6b086fad49", "sources": ["arxiv", "semantic_scholar"], "title": "Federated Orthogonal Training: Mitigating Global Catastrophic Forgetting in Continual Federated Learning", "abstract": "Federated Learning (FL) has gained significant attraction due to its ability to enable privacy-preserving training over decentralized data. Current literature in FL mostly focuses on single-task learning. However, over time, new tasks may appear in the clients and the global model should learn these tasks without forgetting previous tasks. This real-world scenario is known as Continual Federated Learning (CFL). The main challenge of CFL is Global Catastrophic Forgetting, which corresponds to the fact that when the global model is trained on new tasks, its performance on old tasks decreases. There have been a few recent works on CFL to propose methods that aim to address the global catastrophic forgetting problem. However, these works either have unrealistic assumptions on the availability of past data samples or violate the privacy principles of FL. We propose a novel method, Federated Orthogonal Training (FOT), to overcome these drawbacks and address the global catastrophic forgetting in CFL. Our algorithm extracts the global input subspace of each layer for old tasks and modifies the aggregated updates of new tasks such that they are orthogonal to the global principal subspace of old tasks for each layer. This decreases the interference between tasks, which is the main cause for forgetting. We empirically show that FOT outperforms state-of-the-art continual learning methods in the CFL setting, achieving an average accuracy gain of up to 15% with 27% lower forgetting while only incurring a minimal computation and communication cost.", "authors": ["Yavuz Faruk Bakman", "Duygu Nur Yaldiz", "Yahya H. Ezzeldin", "Salman Avestimehr"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2023-09-03", "url": "https://arxiv.org/abs/2309.01289", "pdf_url": "https://arxiv.org/pdf/2309.01289v3", "arxiv_id": "2309.01289", "doi": "10.48550/arXiv.2309.01289", "citation_count": 29, "influential_citation_count": 7, "has_code": false, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.4515}
{"id": "fea12c5dca223a6536e5555e89621135bf9bbe4551a4cd6e92fd2d3ed698e2b5", "sources": ["arxiv", "semantic_scholar"], "title": "Jointly Exploring Client Drift and Catastrophic Forgetting in Dynamic Learning", "abstract": "Federated and Continual Learning have emerged as potential paradigms for the robust and privacy-aware use of Deep Learning in dynamic environments. However, Client Drift and Catastrophic Forgetting are fundamental obstacles to guaranteeing consistent performance. Existing work only addresses these problems separately, which neglects the fact that the root cause behind both forms of performance deterioration is connected. We propose a unified analysis framework for building a controlled test environment for Client Drift -- by perturbing a defined ratio of clients -- and Catastrophic Forgetting -- by shifting all clients with a particular strength. Our framework further leverages this new combined analysis by generating a 3D landscape of the combined performance impact from both. We demonstrate that the performance drop through Client Drift, caused by a certain share of shifted clients, is correlated to the drop from Catastrophic Forgetting resulting from a corresponding shift strength. Correlation tests between both problems for Computer Vision (CelebA) and Medical Imaging (PESO) support this new perspective, with an average Pearson rank correlation coefficient of over 0.94. Our framework's novel ability of combined spatio-temporal shift analysis allows us to investigate how both forms of distribution shift behave in mixed scenarios, opening a new pathway for better generalization. We show that a combination of moderate Client Drift and Catastrophic Forgetting can even improve the performance of the resulting model (causing a \"Generalization Bump\") compared to when only one of the shifts occurs individually. We apply a simple and commonly used method from Continual Learning in the federated setting and observe this phenomenon to be reoccurring, leveraging the ability of our framework to analyze existing and novel methods for Federated and Continual Learning.", "authors": ["Niklas Babendererde", "Moritz Fuchs", "Camila Gonzalez", "Yuri Tolkach", "Anirban Mukhopadhyay"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Medicine", "Computer Science"], "published_date": "2023-09-01", "url": "https://arxiv.org/abs/2309.00688", "pdf_url": "https://arxiv.org/pdf/2309.00688v1", "arxiv_id": "2309.00688", "doi": "10.1038/s41598-025-89873-6", "citation_count": 8, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Scientific Reports", "quality_score": 0.2386}
{"id": "8adbb162e108b7000d8029be9105d1d8d22a523de33ac661b4be7d195cc7d82c", "sources": ["arxiv", "semantic_scholar"], "title": "ScrollNet: Dynamic Weight Importance for Continual Learning", "abstract": "The principle underlying most existing continual learning (CL) methods is to prioritize stability by penalizing changes in parameters crucial to old tasks, while allowing for plasticity in other parameters. The importance of weights for each task can be determined either explicitly through learning a task-specific mask during training (e.g., parameter isolation-based approaches) or implicitly by introducing a regularization term (e.g., regularization-based approaches). However, all these methods assume that the importance of weights for each task is unknown prior to data exposure. In this paper, we propose ScrollNet as a scrolling neural network for continual learning. ScrollNet can be seen as a dynamic network that assigns the ranking of weight importance for each task before data exposure, thus achieving a more favorable stability-plasticity tradeoff during sequential task learning by reassigning this ranking for different tasks. Additionally, we demonstrate that ScrollNet can be combined with various CL methods, including regularization-based and replay-based approaches. Experimental results on CIFAR100 and TinyImagenet datasets show the effectiveness of our proposed method. We release our code at https://github.com/FireFYF/ScrollNet.git.", "authors": ["Fei Yang", "Kai Wang", "Joost van de Weijer"], "categories": ["cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2023-08-31", "url": "https://arxiv.org/abs/2308.16567", "pdf_url": "https://arxiv.org/pdf/2308.16567v1", "arxiv_id": "2308.16567", "doi": "10.1109/ICCVW60793.2023.00359", "citation_count": 7, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/FireFYF/ScrollNet.git", "venue": null, "quality_score": 0.2258}
{"id": "2cb7e97dcd64f52d834c5f9299dc20f94f4c63af272d6876874a55de1d4020e5", "sources": ["arxiv", "semantic_scholar"], "title": "GRASP: A Rehearsal Policy for Efficient Online Continual Learning", "abstract": "Continual learning (CL) in deep neural networks (DNNs) involves incrementally accumulating knowledge in a DNN from a growing data stream. A major challenge in CL is that non-stationary data streams cause catastrophic forgetting of previously learned abilities. A popular solution is rehearsal: storing past observations in a buffer and then sampling the buffer to update the DNN. Uniform sampling in a class-balanced manner is highly effective, and better sample selection policies have been elusive. Here, we propose a new sample selection policy called GRASP that selects the most prototypical (easy) samples first and then gradually selects less prototypical (harder) examples. GRASP has little additional compute or memory overhead compared to uniform selection, enabling it to scale to large datasets. Compared to 17 other rehearsal policies, GRASP achieves higher accuracy in CL experiments on ImageNet. Compared to uniform balanced sampling, GRASP achieves the same performance with 40% fewer updates. We also show that GRASP is effective for CL on five text classification datasets.", "authors": ["Md Yousuf Harun", "Jhair Gallardo", "Junyu Chen", "Christopher Kanan"], "categories": ["cs.LG", "cs.CL", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2023-08-25", "url": "https://arxiv.org/abs/2308.13646", "pdf_url": "https://arxiv.org/pdf/2308.13646v2", "arxiv_id": "2308.13646", "doi": "10.48550/arXiv.2308.13646", "citation_count": 15, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.301}
{"id": "a514a5fdf453736b7ba8af9b63c19ed066448c5e7f6bfe70a148316a1d9510ba", "sources": ["arxiv", "semantic_scholar"], "title": "A Comprehensive Empirical Evaluation on Online Continual Learning", "abstract": "Online continual learning aims to get closer to a live learning experience by learning directly on a stream of data with temporally shifting distribution and by storing a minimum amount of data from that stream. In this empirical evaluation, we evaluate various methods from the literature that tackle online continual learning. More specifically, we focus on the class-incremental setting in the context of image classification, where the learner must learn new classes incrementally from a stream of data. We compare these methods on the Split-CIFAR100 and Split-TinyImagenet benchmarks, and measure their average accuracy, forgetting, stability, and quality of the representations, to evaluate various aspects of the algorithm at the end but also during the whole training period. We find that most methods suffer from stability and underfitting issues. However, the learned representations are comparable to i.i.d. training under the same computational budget. No clear winner emerges from the results and basic experience replay, when properly tuned and implemented, is a very strong baseline. We release our modular and extensible codebase at https://github.com/AlbinSou/ocl_survey based on the avalanche framework to reproduce our results and encourage future research.", "authors": ["Albin Soutif--Cormerais", "Antonio Carta", "Andrea Cossu", "Julio Hurtado", "Hamed Hemati", "Vincenzo Lomonaco", "Joost Van de Weijer"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2023-08-20", "url": "https://arxiv.org/abs/2308.10328", "pdf_url": "https://arxiv.org/pdf/2308.10328v3", "arxiv_id": "2308.10328", "doi": "10.1109/ICCVW60793.2023.00378", "citation_count": 40, "influential_citation_count": 5, "has_code": true, "code_url": "https://github.com/AlbinSou/ocl_survey", "venue": null, "quality_score": 0.4032}
{"id": "988862ef2b7e5936bec495d990fdcf3e2ced9e5129715113589b812b4d194e33", "sources": ["arxiv", "semantic_scholar"], "title": "On the Effectiveness of LayerNorm Tuning for Continual Learning in Vision Transformers", "abstract": "State-of-the-art rehearsal-free continual learning methods exploit the peculiarities of Vision Transformers to learn task-specific prompts, drastically reducing catastrophic forgetting. However, there is a tradeoff between the number of learned parameters and the performance, making such models computationally expensive. In this work, we aim to reduce this cost while maintaining competitive performance. We achieve this by revisiting and extending a simple transfer learning idea: learning task-specific normalization layers. Specifically, we tune the scale and bias parameters of LayerNorm for each continual learning task, selecting them at inference time based on the similarity between task-specific keys and the output of the pre-trained model. To make the classifier robust to incorrect selection of parameters during inference, we introduce a two-stage training procedure, where we first optimize the task-specific parameters and then train the classifier with the same selection procedure of the inference time. Experiments on ImageNet-R and CIFAR-100 show that our method achieves results that are either superior or on par with {the state of the art} while being computationally cheaper.", "authors": ["Thomas De Min", "Massimiliano Mancini", "Karteek Alahari", "Xavier Alameda-Pineda", "Elisa Ricci"], "categories": ["cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2023-08-18", "url": "https://arxiv.org/abs/2308.09610", "pdf_url": "https://arxiv.org/pdf/2308.09610v1", "arxiv_id": "2308.09610", "doi": "10.1109/ICCVW60793.2023.00385", "citation_count": 17, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.3138}
{"id": "46dfc334977a2ead25ca4918726f29e350ba0221805215becdae04918cd0693c", "sources": ["arxiv", "semantic_scholar"], "title": "AdaER: An Adaptive Experience Replay Approach for Continual Lifelong Learning", "abstract": "Continual lifelong learning is an machine learning framework inspired by human learning, where learners are trained to continuously acquire new knowledge in a sequential manner. However, the non-stationary nature of streaming training data poses a significant challenge known as catastrophic forgetting, which refers to the rapid forgetting of previously learned knowledge when new tasks are introduced. While some approaches, such as experience replay (ER), have been proposed to mitigate this issue, their performance remains limited, particularly in the class-incremental scenario which is considered natural and highly challenging. In this paper, we present a novel algorithm, called adaptive-experience replay (AdaER), to address the challenge of continual lifelong learning. AdaER consists of two stages: memory replay and memory update. In the memory replay stage, AdaER introduces a contextually-cued memory recall (C-CMR) strategy, which selectively replays memories that are most conflicting with the current input data in terms of both data and task. Additionally, AdaER incorporates an entropy-balanced reservoir sampling (E-BRS) strategy to enhance the performance of the memory buffer by maximizing information entropy. To evaluate the effectiveness of AdaER, we conduct experiments on established supervised continual lifelong learning benchmarks, specifically focusing on class-incremental learning scenarios. The results demonstrate that AdaER outperforms existing continual lifelong learning baselines, highlighting its efficacy in mitigating catastrophic forgetting and improving learning performance.", "authors": ["Xingyu Li", "Bo Tang", "Haifeng Li"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2023-08-07", "url": "https://arxiv.org/abs/2308.03810", "pdf_url": "https://arxiv.org/pdf/2308.03810v2", "arxiv_id": "2308.03810", "doi": "10.1016/j.neucom.2023.127204", "citation_count": 49, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "Neurocomputing", "quality_score": 0.4247}
{"id": "9f50ed6c3be865f05e9ec0bc4cf8a285118211132412fb611c311b5ac1fc74c4", "sources": ["arxiv", "semantic_scholar"], "title": "Graph Embedded Intuitionistic Fuzzy Random Vector Functional Link Neural Network for Class Imbalance Learning", "abstract": "The domain of machine learning is confronted with a crucial research area known as class imbalance learning, which presents considerable hurdles in precise classification of minority classes. This issue can result in biased models where the majority class takes precedence in the training process, leading to the underrepresentation of the minority class. The random vector functional link (RVFL) network is a widely used and effective learning model for classification due to its good generalization performance and efficiency. However, it suffers when dealing with imbalanced datasets. To overcome this limitation, we propose a novel graph embedded intuitionistic fuzzy RVFL for class imbalance learning (GE-IFRVFL-CIL) model incorporating a weighting mechanism to handle imbalanced datasets. The proposed GE-IFRVFL-CIL model offers plethora of benefits: $(i)$ leveraging graph embedding to preserve the inherent topological structure of the datasets, $(ii)$ employing intuitionistic fuzzy theory to handle uncertainty and imprecision in the data, $(iii)$ and the most important, it tackles class imbalance learning. The amalgamation of a weighting scheme, graph embedding, and intuitionistic fuzzy sets leads to the superior performance of the proposed models on KEEL benchmark imbalanced datasets with and without Gaussian noise. Furthermore, we implemented the proposed GE-IFRVFL-CIL on the ADNI dataset and achieved promising results, demonstrating the model's effectiveness in real-world applications. The proposed GE-IFRVFL-CIL model offers a promising solution to address the class imbalance issue, mitigates the detrimental effect of noise and outliers, and preserves the inherent geometrical structures of the dataset.", "authors": ["M. A. Ganaie", "M. Sajid", "A. K. Malik", "M. Tanveer"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science", "Medicine"], "published_date": "2023-07-15", "url": "https://arxiv.org/abs/2307.07881", "pdf_url": "https://arxiv.org/pdf/2307.07881v2", "arxiv_id": "2307.07881", "doi": "10.1109/TNNLS.2024.3353531", "citation_count": 29, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.3693}
{"id": "672e3fa38baee785d2e2c27fdd220a97fa4e91850b469a6887dfa002fcb5313f", "sources": ["arxiv", "semantic_scholar"], "title": "Learning Active Subspaces and Discovering Important Features with Gaussian Radial Basis Functions Neural Networks", "abstract": "Providing a model that achieves a strong predictive performance and is simultaneously interpretable by humans is one of the most difficult challenges in machine learning research due to the conflicting nature of these two objectives. To address this challenge, we propose a modification of the radial basis function neural network model by equipping its Gaussian kernel with a learnable precision matrix. We show that precious information is contained in the spectrum of the precision matrix that can be extracted once the training of the model is completed. In particular, the eigenvectors explain the directions of maximum sensitivity of the model revealing the active subspace and suggesting potential applications for supervised dimensionality reduction. At the same time, the eigenvectors highlight the relationship in terms of absolute variation between the input and the latent variables, thereby allowing us to extract a ranking of the input variables based on their importance to the prediction task enhancing the model interpretability. We conducted numerical experiments for regression, classification, and feature selection tasks, comparing our model against popular machine learning models, the state-of-the-art deep learning-based embedding feature selection techniques, and a transformer model for tabular data. Our results demonstrate that the proposed model does not only yield an attractive prediction performance compared to the competitors but also provides meaningful and interpretable results that potentially could assist the decision-making process in real-world applications. A PyTorch implementation of the model is available on GitHub at the following link. https://github.com/dannyzx/Gaussian-RBFNN", "authors": ["Danny D'Agostino", "Ilija Ilievski", "Christine Annette Shoemaker"], "categories": ["cs.LG", "cs.AI", "cs.NE", "stat.ML"], "fields_of_study": ["Computer Science", "Medicine", "Mathematics"], "published_date": "2023-07-11", "url": "https://arxiv.org/abs/2307.05639", "pdf_url": "https://arxiv.org/pdf/2307.05639v2", "arxiv_id": "2307.05639", "doi": "10.1016/j.neunet.2024.106335", "citation_count": 12, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/dannyzx/Gaussian-RBFNN", "venue": "Neural Networks", "quality_score": 0.2785}
{"id": "df9b7803ecc667401b63dda5ffa317e4c7238ca69da25276ae682b551c0653c9", "sources": ["arxiv", "semantic_scholar"], "title": "Neural Hilbert Ladders: Multi-Layer Neural Networks in Function Space", "abstract": "To characterize the function space explored by neural networks (NNs) is an important aspect of learning theory. In this work, noticing that a multi-layer NN generates implicitly a hierarchy of reproducing kernel Hilbert spaces (RKHSs) - named a neural Hilbert ladder (NHL) - we define the function space as an infinite union of RKHSs, which generalizes the existing Barron space theory of two-layer NNs. We then establish several theoretical properties of the new space. First, we prove a correspondence between functions expressed by L-layer NNs and those belonging to L-level NHLs. Second, we prove generalization guarantees for learning an NHL with a controlled complexity measure. Third, we derive a non-Markovian dynamics of random fields that governs the evolution of the NHL which is induced by the training of multi-layer NNs in an infinite-width mean-field limit. Fourth, we show examples of depth separation in NHLs under the ReLU activation function. Finally, we perform numerical experiments to illustrate the feature learning aspect of NN training through the lens of NHLs.", "authors": ["Zhengdao Chen"], "categories": ["cs.LG", "math.FA", "math.OC", "math.PR", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2023-07-03", "url": "https://arxiv.org/abs/2307.01177", "pdf_url": "https://arxiv.org/pdf/2307.01177v2", "arxiv_id": "2307.01177", "doi": "10.48550/arXiv.2307.01177", "citation_count": 5, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Journal of machine learning research", "quality_score": 0.1945}
{"id": "26f61b5174ebae0f04ef08b65c0b15b544d1df07e308facb3241817c72dd79aa", "sources": ["arxiv", "semantic_scholar"], "title": "Multi-Fidelity Active Learning with GFlowNets", "abstract": "In the last decades, the capacity to generate large amounts of data in science and engineering applications has been growing steadily. Meanwhile, machine learning has progressed to become a suitable tool to process and utilise the available data. Nonetheless, many relevant scientific and engineering problems present challenges where current machine learning methods cannot yet efficiently leverage the available data and resources. For example, in scientific discovery, we are often faced with the problem of exploring very large, structured and high-dimensional spaces. Moreover, the high fidelity, black-box objective function is often very expensive to evaluate. Progress in machine learning methods that can efficiently tackle such challenges would help accelerate currently crucial areas such as drug and materials discovery. In this paper, we propose a multi-fidelity active learning algorithm with GFlowNets as a sampler, to efficiently discover diverse, high-scoring candidates where multiple approximations of the black-box function are available at lower fidelity and cost. Our evaluation on molecular discovery tasks shows that multi-fidelity active learning with GFlowNets can discover high-scoring candidates at a fraction of the budget of its single-fidelity counterpart while maintaining diversity, unlike RL-based alternatives. These results open new avenues for multi-fidelity active learning to accelerate scientific discovery and engineering design.", "authors": ["Alex Hernandez-Garcia", "Nikita Saxena", "Moksh Jain", "Cheng-Hao Liu", "Yoshua Bengio"], "categories": ["cs.LG", "q-bio.BM"], "fields_of_study": ["Computer Science", "Biology"], "published_date": "2023-06-20", "url": "https://arxiv.org/abs/2306.11715", "pdf_url": "https://arxiv.org/pdf/2306.11715v2", "arxiv_id": "2306.11715", "doi": "10.48550/arXiv.2306.11715", "citation_count": 20, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Transactions on Machine Learning Research (TMLR) 07/2024 https://openreview.net/forum?id=dLaazW9zuF", "quality_score": 0.3306}
{"id": "571c525c34528385373f77a6bb7057cbef9a67f50c621b834352eb23cbaa6960", "sources": ["arxiv", "semantic_scholar"], "title": "Partial Hypernetworks for Continual Learning", "abstract": "Hypernetworks mitigate forgetting in continual learning (CL) by generating task-dependent weights and penalizing weight changes at a meta-model level. Unfortunately, generating all weights is not only computationally expensive for larger architectures, but also, it is not well understood whether generating all model weights is necessary. Inspired by latent replay methods in CL, we propose partial weight generation for the final layers of a model using hypernetworks while freezing the initial layers. With this objective, we first answer the question of how many layers can be frozen without compromising the final performance. Through several experiments, we empirically show that the number of layers that can be frozen is proportional to the distributional similarity in the CL stream. Then, to demonstrate the effectiveness of hypernetworks, we show that noisy streams can significantly impact the performance of latent replay methods, leading to increased forgetting when features from noisy experiences are replayed with old samples. In contrast, partial hypernetworks are more robust to noise by maintaining accuracy on previous experiences. Finally, we conduct experiments on the split CIFAR-100 and TinyImagenet benchmarks and compare different versions of partial hypernetworks to latent replay methods. We conclude that partial weight generation using hypernetworks is a promising solution to the problem of forgetting in neural networks. It can provide an effective balance between computation and final test accuracy in CL streams.", "authors": ["Hamed Hemati", "Vincenzo Lomonaco", "Davide Bacciu", "Damian Borth"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2023-06-19", "url": "https://arxiv.org/abs/2306.10724", "pdf_url": "https://arxiv.org/pdf/2306.10724v1", "arxiv_id": "2306.10724", "doi": "10.48550/arXiv.2306.10724", "citation_count": 11, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2698}
{"id": "cb706baeb27f13a1ea14d0a886bcc780c8f9b462ebb8b0b90cbeab7587e1d9f8", "sources": ["arxiv", "semantic_scholar"], "title": "Continual Adaptation of Vision Transformers for Federated Learning", "abstract": "In this paper, we focus on the important yet understudied problem of Continual Federated Learning (CFL), where a server communicates with a set of clients to incrementally learn new concepts over time without sharing or storing any data. The complexity of this problem is compounded by challenges from both the Continual and Federated Learning perspectives. Specifically, models trained in a CFL setup suffer from catastrophic forgetting which is exacerbated by data heterogeneity across clients. Existing attempts at this problem tend to impose large overheads on clients and communication channels or require access to stored data which renders them unsuitable for real-world use due to privacy. In this paper, we attempt to tackle forgetting and heterogeneity while minimizing overhead costs and without requiring access to any stored data. We study this problem in the context of Vision Transformers and explore parameter-efficient approaches to adapt to dynamic distributions while minimizing forgetting. We achieve this by leveraging a prompting based approach (such that only prompts and classifier heads have to be communicated) and proposing a novel and lightweight generation and distillation scheme to consolidate client models at the server. We formulate this problem for image classification and establish strong baselines for comparison, conduct experiments on CIFAR-100 as well as challenging, large-scale datasets like ImageNet-R and DomainNet. Our approach outperforms both existing methods and our own baselines by as much as 7% while significantly reducing communication and client-level computation costs. Code available at https://github.com/shaunak27/hepco-fed.", "authors": ["Shaunak Halbe", "James Seale Smith", "Junjiao Tian", "Zsolt Kira"], "categories": ["cs.CV", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2023-06-16", "url": "https://arxiv.org/abs/2306.09970", "pdf_url": "https://arxiv.org/pdf/2306.09970v2", "arxiv_id": "2306.09970", "doi": null, "citation_count": 7, "influential_citation_count": 1, "has_code": true, "code_url": "https://github.com/shaunak27/hepco-fed", "venue": null, "quality_score": 0.2258}
{"id": "9e11b9ccd96ecbd3fbee452084c4f6e7a5d960da6ffe266605f76ee5d9615bdb", "sources": ["arxiv", "semantic_scholar"], "title": "Changing Data Sources in the Age of Machine Learning for Official Statistics", "abstract": "Data science has become increasingly essential for the production of official statistics, as it enables the automated collection, processing, and analysis of large amounts of data. With such data science practices in place, it enables more timely, more insightful and more flexible reporting. However, the quality and integrity of data-science-driven statistics rely on the accuracy and reliability of the data sources and the machine learning techniques that support them. In particular, changes in data sources are inevitable to occur and pose significant risks that are crucial to address in the context of machine learning for official statistics. This paper gives an overview of the main risks, liabilities, and uncertainties associated with changing data sources in the context of machine learning for official statistics. We provide a checklist of the most prevalent origins and causes of changing data sources; not only on a technical level but also regarding ownership, ethics, regulation, and public perception. Next, we highlight the repercussions of changing data sources on statistical reporting. These include technical effects such as concept drift, bias, availability, validity, accuracy and completeness, but also the neutrality and potential discontinuation of the statistical offering. We offer a few important precautionary measures, such as enhancing robustness in both data sourcing and statistical techniques, and thorough monitoring. In doing so, machine learning-based official statistics can maintain integrity, reliability, consistency, and relevance in policy-making, decision-making, and public discourse.", "authors": ["Cedric De Boom", "Michael Reusens"], "categories": ["stat.ML", "cs.LG"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2023-06-07", "url": "https://arxiv.org/abs/2306.04338", "pdf_url": "https://arxiv.org/pdf/2306.04338v1", "arxiv_id": "2306.04338", "doi": "10.48550/arXiv.2306.04338", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.0753}
{"id": "61f95cd49e4c25329346847baa5e4c5a790c0a6da6d71fdb32c7b3230fab84cb", "sources": ["arxiv", "semantic_scholar"], "title": "Forgettable Federated Linear Learning with Certified Data Unlearning", "abstract": "Federated Learning (FL) enables collaborative model training across distributed clients while preserving user privacy. Recently, Federated Unlearning (FU) has emerged to address the \"right to be forgotten\" and to remove the influence of poisoned or target clients without retraining the entire FL system. However, many FU methods require communication with retained or target clients, introduce additional security risks, or store historical models, limiting their efficiency and practicality. Moreover, most FU methods for deep neural networks (DNNs) lack theoretical certification due to the complexity of nonlinear models and their training dynamics. In this work, we introduce Forgettable Federated Linear Learning, a training and unlearning framework for DNNs. Our approach uses pre-trained models to linearly approximate DNNs and achieve performance comparable to the original networks through Federated Linear Training. We further present a certified, efficient, and secure unlearning strategy that enables the server to remove a target client's influence without additional client communication or storage. Extensive experiments on small- to large-scale datasets, using both convolutional neural networks and modern foundation models, show that our method balances model accuracy with effective target-client unlearning. This work provides a practical pipeline for efficient and trustworthy FU. Code: https://github.com/Nanboy-Ronan/2F2L-Federated-Unlearning", "authors": ["Ruinan Jin", "Minghui Chen", "Qiong Zhang", "Xiaoxiao Li"], "categories": ["cs.LG", "cs.CV"], "fields_of_study": ["Computer Science", "Medicine"], "published_date": "2023-06-03", "url": "https://arxiv.org/abs/2306.02216", "pdf_url": "https://arxiv.org/pdf/2306.02216v3", "arxiv_id": "2306.02216", "doi": "10.1109/TNNLS.2026.3683398", "citation_count": 7, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/Nanboy-Ronan/2F2L-Federated-Unlearning", "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.2258}
{"id": "50a123238eb0574cc5394de206c835587956c7b37ea88f19672e1a029a2e641b", "sources": ["arxiv", "semantic_scholar"], "title": "Overcoming Catastrophic Forgetting in Massively Multilingual Continual Learning", "abstract": "Real-life multilingual systems should be able to efficiently incorporate new languages as data distributions fed to the system evolve and shift over time. To do this, systems need to handle the issue of catastrophic forgetting, where the model performance drops for languages or tasks seen further in its past. In this paper, we study catastrophic forgetting, as well as methods to minimize this, in a massively multilingual continual learning framework involving up to 51 languages and covering both classification and sequence labeling tasks. We present LR ADJUST, a learning rate scheduling method that is simple, yet effective in preserving new information without strongly overwriting past knowledge. Furthermore, we show that this method is effective across multiple continual learning approaches. Finally, we provide further insights into the dynamics of catastrophic forgetting in this massively multilingual setup.", "authors": ["Genta Indra Winata", "Lingjue Xie", "Karthik Radhakrishnan", "Shijie Wu", "Xisen Jin", "Pengxiang Cheng", "Mayank Kulkarni", "Daniel Preotiuc-Pietro"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2023-05-25", "url": "https://arxiv.org/abs/2305.16252", "pdf_url": "https://arxiv.org/pdf/2305.16252v1", "arxiv_id": "2305.16252", "doi": "10.48550/arXiv.2305.16252", "citation_count": 37, "influential_citation_count": 3, "has_code": false, "code_url": null, "venue": "Annual Meeting of the Association for Computational Linguistics", "quality_score": 0.3949}
{"id": "786814315b2cea8bdcf8cc4f597caacabef537c652ae8fc86660e2057b003890", "sources": ["arxiv", "semantic_scholar"], "title": "Lightweight Learner for Shared Knowledge Lifelong Learning", "abstract": "In Lifelong Learning (LL), agents continually learn as they encounter new conditions and tasks. Most current LL is limited to a single agent that learns tasks sequentially. Dedicated LL machinery is then deployed to mitigate the forgetting of old tasks as new tasks are learned. This is inherently slow. We propose a new Shared Knowledge Lifelong Learning (SKILL) challenge, which deploys a decentralized population of LL agents that each sequentially learn different tasks, with all agents operating independently and in parallel. After learning their respective tasks, agents share and consolidate their knowledge over a decentralized communication network, so that, in the end, all agents can master all tasks. We present one solution to SKILL which uses Lightweight Lifelong Learning (LLL) agents, where the goal is to facilitate efficient sharing by minimizing the fraction of the agent that is specialized for any given task. Each LLL agent thus consists of a common task-agnostic immutable part, where most parameters are, and individual task-specific modules that contain fewer parameters but are adapted to each task. Agents share their task-specific modules, plus summary information (\"task anchors\") representing their tasks in the common task-agnostic latent space of all agents. Receiving agents register each received task-specific module using the corresponding anchor. Thus, every agent improves its ability to solve new tasks each time new task-specific modules and anchors are received. On a new, very challenging SKILL-102 dataset with 102 image classification tasks (5,033 classes in total, 2,041,225 training, 243,464 validation, and 243,464 test images), we achieve much higher (and SOTA) accuracy over 8 LL baselines, while also achieving near perfect parallelization. Code and data can be found at https://github.com/gyhandy/Shared-Knowledge-Lifelong-Learning", "authors": ["Yunhao Ge", "Yuecheng Li", "Di Wu", "Ao Xu", "Adam M. Jones", "Amanda Sofie Rios", "Iordanis Fostiropoulos", "Shixian Wen", "Po-Hsuan Huang", "Zachary William Murdock", "Gozde Sahin", "Shuo Ni", "Kiran Lekkala", "Sumedh Anand Sontakke", "Laurent Itti"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2023-05-24", "url": "https://arxiv.org/abs/2305.15591", "pdf_url": "https://arxiv.org/pdf/2305.15591v1", "arxiv_id": "2305.15591", "doi": "10.48550/arXiv.2305.15591", "citation_count": 14, "influential_citation_count": 2, "has_code": true, "code_url": "https://github.com/gyhandy/Shared-Knowledge-Lifelong-Learning", "venue": null, "quality_score": 0.294}
{"id": "f0a2d09b95807eb8bb57759d032f031b2373d1d7dfd7fe38888df32117481639", "sources": ["arxiv", "semantic_scholar"], "title": "Mitigating Catastrophic Forgetting in Task-Incremental Continual Learning with Adaptive Classification Criterion", "abstract": "Task-incremental continual learning refers to continually training a model in a sequence of tasks while overcoming the problem of catastrophic forgetting (CF). The issue arrives for the reason that the learned representations are forgotten for learning new tasks, and the decision boundary is destructed. Previous studies mostly consider how to recover the representations of learned tasks. It is seldom considered to adapt the decision boundary for new representations and in this paper we propose a Supervised Contrastive learning framework with adaptive classification criterion for Continual Learning (SCCL), In our method, a contrastive loss is used to directly learn representations for different tasks and a limited number of data samples are saved as the classification criterion. During inference, the saved data samples are fed into the current model to obtain updated representations, and a k Nearest Neighbour module is used for classification. In this way, the extensible model can solve the learned tasks with adaptive criteria of saved samples. To mitigate CF, we further use an instance-wise relation distillation regularization term and a memory replay module to maintain the information of previous tasks. Experiments show that SCCL achieves state-of-the-art performance and has a stronger ability to overcome CF compared with the classification baselines.", "authors": ["Yun Luo", "Xiaotian Lin", "Zhen Yang", "Fandong Meng", "Jie Zhou", "Yue Zhang"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2023-05-20", "url": "https://arxiv.org/abs/2305.12270", "pdf_url": "https://arxiv.org/pdf/2305.12270v1", "arxiv_id": "2305.12270", "doi": "10.48550/arXiv.2305.12270", "citation_count": 7, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2258}
{"id": "783d85b955c9cc212a51ba4eabb2ab270b04fdca1cc5ef13d17474f5ec9015e4", "sources": ["arxiv", "semantic_scholar"], "title": "Sharing Lifelong Reinforcement Learning Knowledge via Modulating Masks", "abstract": "Lifelong learning agents aim to learn multiple tasks sequentially over a lifetime. This involves the ability to exploit previous knowledge when learning new tasks and to avoid forgetting. Modulating masks, a specific type of parameter isolation approach, have recently shown promise in both supervised and reinforcement learning. While lifelong learning algorithms have been investigated mainly within a single-agent approach, a question remains on how multiple agents can share lifelong learning knowledge with each other. We show that the parameter isolation mechanism used by modulating masks is particularly suitable for exchanging knowledge among agents in a distributed and decentralized system of lifelong learners. The key idea is that the isolation of specific task knowledge to specific masks allows agents to transfer only specific knowledge on-demand, resulting in robust and effective distributed lifelong learning. We assume fully distributed and asynchronous scenarios with dynamic agent numbers and connectivity. An on-demand communication protocol ensures agents query their peers for specific masks to be transferred and integrated into their policies when facing each task. Experiments indicate that on-demand mask communication is an effective way to implement distributed lifelong reinforcement learning and provides a lifelong learning benefit with respect to distributed RL baselines such as DD-PPO, IMPALA, and PPO+EWC. The system is particularly robust to connection drops and demonstrates rapid learning due to knowledge exchange.", "authors": ["Saptarshi Nath", "Christos Peridis", "Eseoghene Ben-Iwhiwhu", "Xinran Liu", "Shirin Dora", "Cong Liu", "Soheil Kolouri", "Andrea Soltoggio"], "categories": ["cs.LG", "cs.AI", "cs.DC", "cs.MA"], "fields_of_study": ["Computer Science"], "published_date": "2023-05-18", "url": "https://arxiv.org/abs/2305.10997", "pdf_url": "https://arxiv.org/pdf/2305.10997v1", "arxiv_id": "2305.10997", "doi": "10.48550/arXiv.2305.10997", "citation_count": 13, "influential_citation_count": 1, "has_code": true, "code_url": "https://github.com/DMIU-ShELL/deeprl-shell", "venue": null, "quality_score": 0.2865}
{"id": "0ed8cde0c85a61a0ae4163c396d0a0460e6cc804da58320862a257f9af43b65f", "sources": ["arxiv", "semantic_scholar"], "title": "BiRT: Bio-inspired Replay in Vision Transformers for Continual Learning", "abstract": "The ability of deep neural networks to continually learn and adapt to a sequence of tasks has remained challenging due to catastrophic forgetting of previously learned tasks. Humans, on the other hand, have a remarkable ability to acquire, assimilate, and transfer knowledge across tasks throughout their lifetime without catastrophic forgetting. The versatility of the brain can be attributed to the rehearsal of abstract experiences through a complementary learning system. However, representation rehearsal in vision transformers lacks diversity, resulting in overfitting and consequently, performance drops significantly compared to raw image rehearsal. Therefore, we propose BiRT, a novel representation rehearsal-based continual learning approach using vision transformers. Specifically, we introduce constructive noises at various stages of the vision transformer and enforce consistency in predictions with respect to an exponential moving average of the working model. Our method provides consistent performance gain over raw image and vanilla representation rehearsal on several challenging CL benchmarks, while being memory efficient and robust to natural and adversarial corruptions.", "authors": ["Kishaan Jeeveswaran", "Prashant Bhat", "Bahram Zonooz", "Elahe Arani"], "categories": ["cs.CV", "cs.LG", "cs.NE"], "fields_of_study": ["Computer Science"], "published_date": "2023-05-08", "url": "https://arxiv.org/abs/2305.04769", "pdf_url": "https://arxiv.org/pdf/2305.04769v1", "arxiv_id": "2305.04769", "doi": "10.48550/arXiv.2305.04769", "citation_count": 32, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.3796}
{"id": "b0468d8faf1b76eddcdf0d87865e7125eee7d9478276666ac005af1ff9192838", "sources": ["arxiv", "semantic_scholar"], "title": "Continual Reasoning: Non-Monotonic Reasoning in Neurosymbolic AI using Continual Learning", "abstract": "Despite the extensive investment and impressive recent progress at reasoning by similarity, deep learning continues to struggle with more complex forms of reasoning such as non-monotonic and commonsense reasoning. Non-monotonicity is a property of non-classical reasoning typically seen in commonsense reasoning, whereby a reasoning system is allowed (differently from classical logic) to jump to conclusions which may be retracted later, when new information becomes available. Neural-symbolic systems such as Logic Tensor Networks (LTN) have been shown to be effective at enabling deep neural networks to achieve reasoning capabilities. In this paper, we show that by combining a neural-symbolic system with methods from continual learning, LTN can obtain a higher level of accuracy when addressing non-monotonic reasoning tasks. Continual learning is added to LTNs by adopting a curriculum of learning from knowledge and data with recall. We call this process Continual Reasoning, a new methodology for the application of neural-symbolic systems to reasoning tasks. Continual Reasoning is applied to a prototypical non-monotonic reasoning problem as well as other reasoning examples. Experimentation is conducted to compare and analyze the effects that different curriculum choices may have on overall learning and reasoning results. Results indicate significant improvement on the prototypical non-monotonic reasoning problem and a promising outlook for the proposed approach on statistical relational learning examples.", "authors": ["Sofoklis Kyriakopoulos", "Artur S. d'Avila Garcez"], "categories": ["cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2023-05-03", "url": "https://arxiv.org/abs/2305.02171", "pdf_url": "https://arxiv.org/pdf/2305.02171v1", "arxiv_id": "2305.02171", "doi": "10.48550/arXiv.2305.02171", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "International Workshop on Neural-Symbolic Learning and Reasoning", "quality_score": 0.0}
{"id": "89a5570b818e8f407df9c33b689e34a9aa68da2deec514f895e99979e1fd211a", "sources": ["arxiv", "semantic_scholar"], "title": "A Study of Biologically Plausible Neural Network: The Role and Interactions of Brain-Inspired Mechanisms in Continual Learning", "abstract": "Humans excel at continually acquiring, consolidating, and retaining information from an ever-changing environment, whereas artificial neural networks (ANNs) exhibit catastrophic forgetting. There are considerable differences in the complexity of synapses, the processing of information, and the learning mechanisms in biological neural networks and their artificial counterparts, which may explain the mismatch in performance. We consider a biologically plausible framework that constitutes separate populations of exclusively excitatory and inhibitory neurons that adhere to Dale's principle, and the excitatory pyramidal neurons are augmented with dendritic-like structures for context-dependent processing of stimuli. We then conduct a comprehensive study on the role and interactions of different mechanisms inspired by the brain, including sparse non-overlapping representations, Hebbian learning, synaptic consolidation, and replay of past activations that accompanied the learning event. Our study suggests that the employing of multiple complementary mechanisms in a biologically plausible architecture, similar to the brain, may be effective in enabling continual learning in ANNs.", "authors": ["Fahad Sarfraz", "Elahe Arani", "Bahram Zonooz"], "categories": ["cs.NE", "cs.AI", "cs.CV", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2023-04-13", "url": "https://arxiv.org/abs/2304.06738", "pdf_url": "https://arxiv.org/pdf/2304.06738v1", "arxiv_id": "2304.06738", "doi": "10.48550/arXiv.2304.06738", "citation_count": 3, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.1505}
{"id": "662c5b11c5281979eaff6f5ed3e693e69ca8411819621bf86dfe9dcad54166af", "sources": ["arxiv", "semantic_scholar"], "title": "Multi-annotator Deep Learning: A Probabilistic Framework for Classification", "abstract": "Solving complex classification tasks using deep neural networks typically requires large amounts of annotated data. However, corresponding class labels are noisy when provided by error-prone annotators, e.g., crowdworkers. Training standard deep neural networks leads to subpar performances in such multi-annotator supervised learning settings. We address this issue by presenting a probabilistic training framework named multi-annotator deep learning (MaDL). A downstream ground truth and an annotator performance model are jointly trained in an end-to-end learning approach. The ground truth model learns to predict instances' true class labels, while the annotator performance model infers probabilistic estimates of annotators' performances. A modular network architecture enables us to make varying assumptions regarding annotators' performances, e.g., an optional class or instance dependency. Further, we learn annotator embeddings to estimate annotators' densities within a latent space as proxies of their potentially correlated annotations. Together with a weighted loss function, we improve the learning from correlated annotation patterns. In a comprehensive evaluation, we examine three research questions about multi-annotator supervised learning. Our findings show MaDL's state-of-the-art performance and robustness against many correlated, spamming annotators.", "authors": ["Marek Herde", "Denis Huseljic", "Bernhard Sick"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2023-04-05", "url": "https://arxiv.org/abs/2304.02539", "pdf_url": "https://arxiv.org/pdf/2304.02539v2", "arxiv_id": "2304.02539", "doi": "10.48550/arXiv.2304.02539", "citation_count": 17, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "Transactions on Machine Learning Research, 2023", "quality_score": 0.3138}
{"id": "64da758a493ed33fe80ab3bb1d35a431ffa1c750922841e0798a61304dc50d44", "sources": ["arxiv", "semantic_scholar"], "title": "Physics-Inspired Interpretability Of Machine Learning Models", "abstract": "The ability to explain decisions made by machine learning models remains one of the most significant hurdles towards widespread adoption of AI in highly sensitive areas such as medicine, cybersecurity or autonomous driving. Great interest exists in understanding which features of the input data prompt model decision making. In this contribution, we propose a novel approach to identify relevant features of the input data, inspired by methods from the energy landscapes field, developed in the physical sciences. By identifying conserved weights within groups of minima of the loss landscapes, we can identify the drivers of model decision making. Analogues to this idea exist in the molecular sciences, where coordinate invariants or order parameters are employed to identify critical features of a molecule. However, no such approach exists for machine learning loss landscapes. We will demonstrate the applicability of energy landscape methods to machine learning models and give examples, both synthetic and from the real world, for how these methods can help to make models more interpretable.", "authors": ["Maximilian P Niroomand", "David J Wales"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2023-04-05", "url": "https://arxiv.org/abs/2304.02381", "pdf_url": "https://arxiv.org/pdf/2304.02381v2", "arxiv_id": "2304.02381", "doi": "10.48550/arXiv.2304.02381", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.0753}
{"id": "56778ba4a9e880eaff3bd8a7c03588dcbb020cb4d8870509f4dd9d4be95ac014", "sources": ["arxiv", "semantic_scholar"], "title": "Structure Learning with Continuous Optimization: A Sober Look and Beyond", "abstract": "This paper investigates in which cases continuous optimization for directed acyclic graph (DAG) structure learning can and cannot perform well and why this happens, and suggests possible directions to make the search procedure more reliable. Reisach et al. (2021) suggested that the remarkable performance of several continuous structure learning approaches is primarily driven by a high agreement between the order of increasing marginal variances and the topological order, and demonstrated that these approaches do not perform well after data standardization. We analyze this phenomenon for continuous approaches assuming equal and non-equal noise variances, and show that the statement may not hold in either case by providing counterexamples, justifications, and possible alternative explanations. We further demonstrate that nonconvexity may be a main concern especially for the non-equal noise variances formulation, while recent advances in continuous structure learning fail to achieve improvement in this case. Our findings suggest that future works should take into account the non-equal noise variances formulation to handle more general settings and for a more comprehensive empirical evaluation. Lastly, we provide insights into other aspects of the search procedure, including thresholding and sparsity, and show that they play an important role in the final solutions.", "authors": ["Ignavier Ng", "Biwei Huang", "Kun Zhang"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2023-04-04", "url": "https://arxiv.org/abs/2304.02146", "pdf_url": "https://arxiv.org/pdf/2304.02146v2", "arxiv_id": "2304.02146", "doi": "10.48550/arXiv.2304.02146", "citation_count": 37, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": "CLEaR", "quality_score": 0.3949}
{"id": "2250c529e8ed6eb2facfef0826b06cf9e88cce8b0200ee177304495b505fad43", "sources": ["arxiv", "semantic_scholar"], "title": "Online Distillation with Continual Learning for Cyclic Domain Shifts", "abstract": "In recent years, online distillation has emerged as a powerful technique for adapting real-time deep neural networks on the fly using a slow, but accurate teacher model. However, a major challenge in online distillation is catastrophic forgetting when the domain shifts, which occurs when the student model is updated with data from the new domain and forgets previously learned knowledge. In this paper, we propose a solution to this issue by leveraging the power of continual learning methods to reduce the impact of domain shifts. Specifically, we integrate several state-of-the-art continual learning methods in the context of online distillation and demonstrate their effectiveness in reducing catastrophic forgetting. Furthermore, we provide a detailed analysis of our proposed solution in the case of cyclic domain shifts. Our experimental results demonstrate the efficacy of our approach in improving the robustness and accuracy of online distillation, with potential applications in domains such as video surveillance or autonomous driving. Overall, our work represents an important step forward in the field of online distillation and continual learning, with the potential to significantly impact real-world applications.", "authors": ["Joachim Houyon", "Anthony Cioppa", "Yasir Ghunaim", "Motasem Alfarra", "Anaïs Halin", "Maxim Henry", "Bernard Ghanem", "Marc Van Droogenbroeck"], "categories": ["cs.CV", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2023-04-03", "url": "https://arxiv.org/abs/2304.01239", "pdf_url": "https://arxiv.org/pdf/2304.01239v1", "arxiv_id": "2304.01239", "doi": "10.1109/CVPRW59228.2023.00242", "citation_count": 11, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2698}
{"id": "678cd13f8260a9c48434e3e93cf0441079e1b33346c51e62e0a236c4fa101993", "sources": ["arxiv", "semantic_scholar"], "title": "Knowledge Accumulation in Continually Learned Representations and the Issue of Feature Forgetting", "abstract": "Continual learning research has shown that neural networks suffer from catastrophic forgetting \"at the output level\", but it is debated whether this is also the case at the level of learned representations. Multiple recent studies ascribe representations a certain level of innate robustness against forgetting -- that they only forget minimally in comparison with forgetting at the output level. We revisit and expand upon the experiments that revealed this difference in forgetting and illustrate the coexistence of two phenomena that affect the quality of continually learned representations: knowledge accumulation and feature forgetting. Taking both aspects into account, we show that, even though forgetting in the representation (i.e. feature forgetting) can be small in absolute terms, when measuring relative to how much was learned during a task, forgetting in the representation tends to be just as catastrophic as forgetting at the output level. Next we show that this feature forgetting is problematic as it substantially slows down the incremental learning of good general representations (i.e. knowledge accumulation). Finally, we study how feature forgetting and knowledge accumulation are affected by different types of continual learning methods.", "authors": ["Timm Hess", "Eli Verwimp", "Gido M. van de Ven", "Tinne Tuytelaars"], "categories": ["cs.LG", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2023-04-03", "url": "https://arxiv.org/abs/2304.00933", "pdf_url": "https://arxiv.org/pdf/2304.00933v4", "arxiv_id": "2304.00933", "doi": "10.48550/arXiv.2304.00933", "citation_count": 12, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "Transactions on Machine Learning Research (TMLR), 2024", "quality_score": 0.2785}
{"id": "1a70e9842f0cad2d22f8efcb8f37f1b1dfca12ff829caf24e33354499bcbcb9b", "sources": ["arxiv", "semantic_scholar"], "title": "Optimal Goal-Reaching Reinforcement Learning via Quasimetric Learning", "abstract": "In goal-reaching reinforcement learning (RL), the optimal value function has a particular geometry, called quasimetric structure. This paper introduces Quasimetric Reinforcement Learning (QRL), a new RL method that utilizes quasimetric models to learn optimal value functions. Distinct from prior approaches, the QRL objective is specifically designed for quasimetrics, and provides strong theoretical recovery guarantees. Empirically, we conduct thorough analyses on a discretized MountainCar environment, identifying properties of QRL and its advantages over alternatives. On offline and online goal-reaching benchmarks, QRL also demonstrates improved sample efficiency and performance, across both state-based and image-based observations.", "authors": ["Tongzhou Wang", "Antonio Torralba", "Phillip Isola", "Amy Zhang"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2023-04-03", "url": "https://arxiv.org/abs/2304.01203", "pdf_url": "https://arxiv.org/pdf/2304.01203v7", "arxiv_id": "2304.01203", "doi": "10.48550/arXiv.2304.01203", "citation_count": 103, "influential_citation_count": 26, "has_code": true, "code_url": "https://github.com/quasimetric-learning/quasimetric-rl/", "venue": "International Conference on Machine Learning", "quality_score": 0.7157}
{"id": "5dc1fb9db2484a9fe93efda4499e2e99544b7188676ce579876e72034c93dbdd", "sources": ["arxiv", "semantic_scholar"], "title": "How Efficient Are Today's Continual Learning Algorithms?", "abstract": "Supervised Continual learning involves updating a deep neural network (DNN) from an ever-growing stream of labeled data. While most work has focused on overcoming catastrophic forgetting, one of the major motivations behind continual learning is being able to efficiently update a network with new information, rather than retraining from scratch on the training dataset as it grows over time. Despite recent continual learning methods largely solving the catastrophic forgetting problem, there has been little attention paid to the efficiency of these algorithms. Here, we study recent methods for incremental class learning and illustrate that many are highly inefficient in terms of compute, memory, and storage. Some methods even require more compute than training from scratch! We argue that for continual learning to have real-world applicability, the research community cannot ignore the resources used by these algorithms. There is more to continual learning than mitigating catastrophic forgetting.", "authors": ["Md Yousuf Harun", "Jhair Gallardo", "Tyler L. Hayes", "Christopher Kanan"], "categories": ["cs.CV", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2023-03-29", "url": "https://arxiv.org/abs/2303.18171", "pdf_url": "https://arxiv.org/pdf/2303.18171v2", "arxiv_id": "2303.18171", "doi": "10.1109/CVPRW59228.2023.00241", "citation_count": 32, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.3796}
{"id": "e28750862dc20224f1520a8f6f9e7cd55120e90e37588aecef3cff78d53c3dc6", "sources": ["arxiv", "semantic_scholar"], "title": "Privacy-preserving machine learning for healthcare: open challenges and future perspectives", "abstract": "Machine Learning (ML) has recently shown tremendous success in modeling various healthcare prediction tasks, ranging from disease diagnosis and prognosis to patient treatment. Due to the sensitive nature of medical data, privacy must be considered along the entire ML pipeline, from model training to inference. In this paper, we conduct a review of recent literature concerning Privacy-Preserving Machine Learning (PPML) for healthcare. We primarily focus on privacy-preserving training and inference-as-a-service, and perform a comprehensive review of existing trends, identify challenges, and discuss opportunities for future research directions. The aim of this review is to guide the development of private and efficient ML models in healthcare, with the prospects of translating research efforts into real-world settings.", "authors": ["Alejandro Guerra-Manzanares", "L. Julian Lechuga Lopez", "Michail Maniatakos", "Farah E. Shamout"], "categories": ["cs.LG", "cs.CR"], "fields_of_study": ["Computer Science"], "published_date": "2023-03-27", "url": "https://arxiv.org/abs/2303.15563", "pdf_url": "https://arxiv.org/pdf/2303.15563v1", "arxiv_id": "2303.15563", "doi": "10.1007/978-3-031-39539-0_3", "citation_count": 25, "influential_citation_count": 3, "has_code": false, "code_url": null, "venue": "Trustworthy Machine Learning for Healthcare. TML4H 2023. Lecture Notes in Computer Science, vol 13932", "quality_score": 0.3537}
{"id": "e2a28a8236282b456d4f2d5063f0fafd5e2e45f151da3c9fc8e17447434ca1fa", "sources": ["arxiv", "semantic_scholar"], "title": "Learn, Unlearn and Relearn: An Online Learning Paradigm for Deep Neural Networks", "abstract": "Deep neural networks (DNNs) are often trained on the premise that the complete training data set is provided ahead of time. However, in real-world scenarios, data often arrive in chunks over time. This leads to important considerations about the optimal strategy for training DNNs, such as whether to fine-tune them with each chunk of incoming data (warm-start) or to retrain them from scratch with the entire corpus of data whenever a new chunk is available. While employing the latter for training can be resource-intensive, recent work has pointed out the lack of generalization in warm-start models. Therefore, to strike a balance between efficiency and generalization, we introduce Learn, Unlearn, and Relearn (LURE) an online learning paradigm for DNNs. LURE interchanges between the unlearning phase, which selectively forgets the undesirable information in the model through weight reinitialization in a data-dependent manner, and the relearning phase, which emphasizes learning on generalizable features. We show that our training paradigm provides consistent performance gains across datasets in both classification and few-shot settings. We further show that it leads to more robust and well-calibrated models.", "authors": ["Vijaya Raghavan T. Ramkumar", "Elahe Arani", "Bahram Zonooz"], "categories": ["cs.LG", "cs.AI", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2023-03-18", "url": "https://arxiv.org/abs/2303.10455", "pdf_url": "https://arxiv.org/pdf/2303.10455v1", "arxiv_id": "2303.10455", "doi": "10.48550/arXiv.2303.10455", "citation_count": 9, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.25}
{"id": "88a3fc48668fde3300811d5562e98e1342dfd4aae64e58feebcde4e87293e750", "sources": ["arxiv", "semantic_scholar"], "title": "Aux-Drop: Handling Haphazard Inputs in Online Learning Using Auxiliary Dropouts", "abstract": "Many real-world applications based on online learning produce streaming data that is haphazard in nature, i.e., contains missing features, features becoming obsolete in time, the appearance of new features at later points in time and a lack of clarity on the total number of input features. These challenges make it hard to build a learnable system for such applications, and almost no work exists in deep learning that addresses this issue. In this paper, we present Aux-Drop, an auxiliary dropout regularization strategy for online learning that handles the haphazard input features in an effective manner. Aux-Drop adapts the conventional dropout regularization scheme for the haphazard input feature space ensuring that the final output is minimally impacted by the chaotic appearance of such features. It helps to prevent the co-adaptation of especially the auxiliary and base features, as well as reduces the strong dependence of the output on any of the auxiliary inputs of the model. This helps in better learning for scenarios where certain features disappear in time or when new features are to be modelled. The efficacy of Aux-Drop has been demonstrated through extensive numerical experiments on SOTA benchmarking datasets that include Italy Power Demand, HIGGS, SUSY and multiple UCI datasets. The code is available at https://github.com/Rohit102497/Aux-Drop.", "authors": ["Rohit Agarwal", "Deepak Gupta", "Alexander Horsch", "Dilip K. Prasad"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2023-03-09", "url": "https://arxiv.org/abs/2303.05155", "pdf_url": "https://arxiv.org/pdf/2303.05155v2", "arxiv_id": "2303.05155", "doi": "10.48550/arXiv.2303.05155", "citation_count": 5, "influential_citation_count": 1, "has_code": true, "code_url": "https://github.com/Rohit102497/Aux-Drop", "venue": "Transactions on Machine Learning Research, 2023", "quality_score": 0.1945}
{"id": "2850fb7fdcc107ead3047e62fe8c4171803f0bbb32e107b7cd77ce1aefa04521", "sources": ["arxiv", "semantic_scholar"], "title": "Robustness-preserving Lifelong Learning via Dataset Condensation", "abstract": "Lifelong learning (LL) aims to improve a predictive model as the data source evolves continuously. Most work in this learning paradigm has focused on resolving the problem of 'catastrophic forgetting,' which refers to a notorious dilemma between improving model accuracy over new data and retaining accuracy over previous data. Yet, it is also known that machine learning (ML) models can be vulnerable in the sense that tiny, adversarial input perturbations can deceive the models into producing erroneous predictions. This motivates the research objective of this paper - specification of a new LL framework that can salvage model robustness (against adversarial attacks) from catastrophic forgetting. Specifically, we propose a new memory-replay LL strategy that leverages modern bi-level optimization techniques to determine the 'coreset' of the current data (i.e., a small amount of data to be memorized) for ease of preserving adversarial robustness over time. We term the resulting LL framework 'Data-Efficient Robustness-Preserving LL' (DERPLL). The effectiveness of DERPLL is evaluated for class-incremental image classification using ResNet-18 over the CIFAR-10 dataset. Experimental results show that DERPLL outperforms the conventional coreset-guided LL baseline and achieves a substantial improvement in both standard accuracy and robust accuracy.", "authors": ["Jinghan Jia", "Yihua Zhang", "Dogyoon Song", "Sijia Liu", "Alfred Hero"], "categories": ["cs.LG", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2023-03-07", "url": "https://arxiv.org/abs/2303.04183", "pdf_url": "https://arxiv.org/pdf/2303.04183v1", "arxiv_id": "2303.04183", "doi": "10.1109/ICASSP49357.2023.10096756", "citation_count": 5, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "IEEE International Conference on Acoustics, Speech, and Signal Processing", "quality_score": 0.1945}
{"id": "e0a9ff3afd153bb3dadf674bd70f2f1c4568d84e9c8efb4654519ac138eb10fc", "sources": ["arxiv", "semantic_scholar"], "title": "Spectral learning of Bernoulli linear dynamical systems models", "abstract": "Latent linear dynamical systems with Bernoulli observations provide a powerful modeling framework for identifying the temporal dynamics underlying binary time series data, which arise in a variety of contexts such as binary decision-making and discrete stochastic processes (e.g., binned neural spike trains). Here we develop a spectral learning method for fast, efficient fitting of probit-Bernoulli latent linear dynamical system (LDS) models. Our approach extends traditional subspace identification methods to the Bernoulli setting via a transformation of the first and second sample moments. This results in a robust, fixed-cost estimator that avoids the hazards of local optima and the long computation time of iterative fitting procedures like the expectation-maximization (EM) algorithm. In regimes where data is limited or assumptions about the statistical structure of the data are not met, we demonstrate that the spectral estimate provides a good initialization for Laplace-EM fitting. Finally, we show that the estimator provides substantial benefits to real world settings by analyzing data from mice performing a sensory decision-making task.", "authors": ["Iris R. Stone", "Yotam Sagiv", "Il Memming Park", "Jonathan W. Pillow"], "categories": ["stat.ML", "cs.LG"], "fields_of_study": ["Medicine", "Mathematics", "Computer Science"], "published_date": "2023-03-03", "url": "https://arxiv.org/abs/2303.02060", "pdf_url": "https://arxiv.org/pdf/2303.02060v2", "arxiv_id": "2303.02060", "doi": "10.48550/arXiv.2303.02060", "citation_count": 2, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Transactions on Machine Learning Research (2023)", "quality_score": 0.1193}
{"id": "22d54f83d3a4684d6e4fff3f08a549c0b28176959b1a41f57f883f73610362e9", "sources": ["arxiv", "semantic_scholar"], "title": "Active learning for data streams: a survey", "abstract": "Online active learning is a paradigm in machine learning that aims to select the most informative data points to label from a data stream. The problem of minimizing the cost associated with collecting labeled observations has gained a lot of attention in recent years, particularly in real-world applications where data is only available in an unlabeled form. Annotating each observation can be time-consuming and costly, making it difficult to obtain large amounts of labeled data. To overcome this issue, many active learning strategies have been proposed in the last decades, aiming to select the most informative observations for labeling in order to improve the performance of machine learning models. These approaches can be broadly divided into two categories: static pool-based and stream-based active learning. Pool-based active learning involves selecting a subset of observations from a closed pool of unlabeled data, and it has been the focus of many surveys and literature reviews. However, the growing availability of data streams has led to an increase in the number of approaches that focus on online active learning, which involves continuously selecting and labeling observations as they arrive in a stream. This work aims to provide an overview of the most recently proposed approaches for selecting the most informative observations from data streams in real time. We review the various techniques that have been proposed and discuss their strengths and limitations, as well as the challenges and opportunities that exist in this area of research.", "authors": ["Davide Cacciarelli", "Murat Kulahci"], "categories": ["stat.ML", "cs.LG", "stat.ME"], "fields_of_study": ["Mathematics", "Computer Science"], "published_date": "2023-02-17", "url": "https://arxiv.org/abs/2302.08893", "pdf_url": "https://arxiv.org/pdf/2302.08893v4", "arxiv_id": "2302.08893", "doi": "10.1007/s10994-023-06454-2", "citation_count": 102, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "Machine-mediated learning", "quality_score": 0.5032}
{"id": "80e16597d39af80d57d1f523fcfbf277c7602e3e6fee7cebf32206e0cc648b0b", "sources": ["arxiv", "semantic_scholar"], "title": "On a continuous time model of gradient descent dynamics and instability in deep learning", "abstract": "The recipe behind the success of deep learning has been the combination of neural networks and gradient-based optimization. Understanding the behavior of gradient descent however, and particularly its instability, has lagged behind its empirical success. To add to the theoretical tools available to study gradient descent we propose the principal flow (PF), a continuous time flow that approximates gradient descent dynamics. To our knowledge, the PF is the only continuous flow that captures the divergent and oscillatory behaviors of gradient descent, including escaping local minima and saddle points. Through its dependence on the eigendecomposition of the Hessian the PF sheds light on the recently observed edge of stability phenomena in deep learning. Using our new understanding of instability we propose a learning rate adaptation method which enables us to control the trade-off between training stability and test set evaluation performance.", "authors": ["Mihaela Rosca", "Yan Wu", "Chongli Qin", "Benoit Dherin"], "categories": ["stat.ML", "cs.LG"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2023-02-03", "url": "https://arxiv.org/abs/2302.01952", "pdf_url": "https://arxiv.org/pdf/2302.01952v3", "arxiv_id": "2302.01952", "doi": "10.48550/arXiv.2302.01952", "citation_count": 16, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.3076}
{"id": "d3c73c6667f18b00dc4a72187c95249ac8b119b31d77f184700b0bbf91057b2b", "sources": ["arxiv", "semantic_scholar"], "title": "Neuro-Symbolic Continual Learning: Knowledge, Reasoning Shortcuts and Concept Rehearsal", "abstract": "We introduce Neuro-Symbolic Continual Learning, where a model has to solve a sequence of neuro-symbolic tasks, that is, it has to map sub-symbolic inputs to high-level concepts and compute predictions by reasoning consistently with prior knowledge. Our key observation is that neuro-symbolic tasks, although different, often share concepts whose semantics remains stable over time. Traditional approaches fall short: existing continual strategies ignore knowledge altogether, while stock neuro-symbolic architectures suffer from catastrophic forgetting. We show that leveraging prior knowledge by combining neuro-symbolic architectures with continual strategies does help avoid catastrophic forgetting, but also that doing so can yield models affected by reasoning shortcuts. These undermine the semantics of the acquired concepts, even when detailed prior knowledge is provided upfront and inference is exact, and in turn continual performance. To overcome these issues, we introduce COOL, a COncept-level cOntinual Learning strategy tailored for neuro-symbolic continual problems that acquires high-quality concepts and remembers them over time. Our experiments on three novel benchmarks highlights how COOL attains sustained high performance on neuro-symbolic continual learning tasks in which other strategies fail.", "authors": ["Emanuele Marconato", "Gianpaolo Bontempo", "Elisa Ficarra", "Simone Calderara", "Andrea Passerini", "Stefano Teso"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2023-02-02", "url": "https://arxiv.org/abs/2302.01242", "pdf_url": "https://arxiv.org/pdf/2302.01242v2", "arxiv_id": "2302.01242", "doi": "10.48550/arXiv.2302.01242", "citation_count": 35, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.3891}
{"id": "8e9875b9cbe0958ea1d7e521728498341f0bb8b3f7ba222cd52baa7a45b0ff45", "sources": ["arxiv", "semantic_scholar"], "title": "Exploring Image Augmentations for Siamese Representation Learning with Chest X-Rays", "abstract": "Image augmentations are quintessential for effective visual representation learning across self-supervised learning techniques. While augmentation strategies for natural imaging have been studied extensively, medical images are vastly different from their natural counterparts. Thus, it is unknown whether common augmentation strategies employed in Siamese representation learning generalize to medical images and to what extent. To address this challenge, in this study, we systematically assess the effect of various augmentations on the quality and robustness of the learned representations. We train and evaluate Siamese Networks for abnormality detection on chest X-Rays across three large datasets (MIMIC-CXR, CheXpert and VinDR-CXR). We investigate the efficacy of the learned representations through experiments involving linear probing, fine-tuning, zero-shot transfer, and data efficiency. Finally, we identify a set of augmentations that yield robust representations that generalize well to both out-of-distribution data and diseases, while outperforming supervised baselines using just zero-shot transfer and linear probes by up to 20%. Our code is available at https://github.com/StanfordMIMI/siaug.", "authors": ["Rogier van der Sluijs", "Nandita Bhaskhar", "Daniel Rubin", "Curtis Langlotz", "Akshay Chaudhari"], "categories": ["eess.IV", "cs.AI", "cs.CV", "cs.LG"], "fields_of_study": ["Computer Science", "Engineering"], "published_date": "2023-01-30", "url": "https://arxiv.org/abs/2301.12636", "pdf_url": "https://arxiv.org/pdf/2301.12636v2", "arxiv_id": "2301.12636", "doi": "10.48550/arXiv.2301.12636", "citation_count": 19, "influential_citation_count": 1, "has_code": true, "code_url": "https://github.com/StanfordMIMI/siaug", "venue": "International Conference on Medical Imaging with Deep Learning", "quality_score": 0.3253}
{"id": "cbc86cebfabde560287838b7824c589925e3ce4b9fa3cd6b87d262ebfbaebf03", "sources": ["arxiv", "semantic_scholar"], "title": "Scalable Real-Time Recurrent Learning Using Columnar-Constructive Networks", "abstract": "Constructing states from sequences of observations is an important component of reinforcement learning agents. One solution for state construction is to use recurrent neural networks. Back-propagation through time (BPTT), and real-time recurrent learning (RTRL) are two popular gradient-based methods for recurrent learning. BPTT requires complete trajectories of observations before it can compute the gradients and is unsuitable for online updates. RTRL can do online updates but scales poorly to large networks. In this paper, we propose two constraints that make RTRL scalable. We show that by either decomposing the network into independent modules or learning the network in stages, we can make RTRL scale linearly with the number of parameters. Unlike prior scalable gradient estimation algorithms, such as UORO and Truncated-BPTT, our algorithms do not add noise or bias to the gradient estimate. Instead, they trade off the functional capacity of the network for computationally efficient learning. We demonstrate the effectiveness of our approach over Truncated-BPTT on a prediction benchmark inspired by animal learning and by doing policy evaluation of pre-trained policies for Atari 2600 games.", "authors": ["Khurram Javed", "Haseeb Shah", "Rich Sutton", "Martha White"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2023-01-20", "url": "https://arxiv.org/abs/2302.05326", "pdf_url": "https://arxiv.org/pdf/2302.05326v3", "arxiv_id": "2302.05326", "doi": null, "citation_count": 15, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Journal of machine learning research", "quality_score": 0.301}
{"id": "c3800081c37cf66b2eb69b48327a432cd64ca0bddb5ed2c83a1c4d1ca51641e5", "sources": ["arxiv", "semantic_scholar"], "title": "A Tutorial on Meta-Reinforcement Learning", "abstract": "While deep reinforcement learning (RL) has fueled multiple high-profile successes in machine learning, it is held back from more widespread adoption by its often poor data efficiency and the limited generality of the policies it produces. A promising approach for alleviating these limitations is to cast the development of better RL algorithms as a machine learning problem itself in a process called meta-RL. Meta-RL is most commonly studied in a problem setting where, given a distribution of tasks, the goal is to learn a policy that is capable of adapting to any new task from the task distribution with as little data as possible. In this survey, we describe the meta-RL problem setting in detail as well as its major variations. We discuss how, at a high level, meta-RL research can be clustered based on the presence of a task distribution and the learning budget available for each individual task. Using these clusters, we then survey meta-RL algorithms and applications. We conclude by presenting the open problems on the path to making meta-RL part of the standard toolbox for a deep RL practitioner.", "authors": ["Jacob Beck", "Risto Vuorio", "Evan Zheran Liu", "Zheng Xiong", "Luisa Zintgraf", "Chelsea Finn", "Shimon Whiteson"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2023-01-19", "url": "https://arxiv.org/abs/2301.08028", "pdf_url": "https://arxiv.org/pdf/2301.08028v4", "arxiv_id": "2301.08028", "doi": "10.1561/2200000080", "citation_count": 158, "influential_citation_count": 8, "has_code": false, "code_url": null, "venue": "Foundations and Trends in Machine Learning: Vol. 18, No. 2-3, pp 224-384 (2025)", "quality_score": 0.5503}
{"id": "87b34ff68482e8c2db20796550ed59c12d222a9eb9f94e955f4a65df9f888313", "sources": ["arxiv", "semantic_scholar"], "title": "Self-Activating Neural Ensembles for Continual Reinforcement Learning", "abstract": "The ability for an agent to continuously learn new skills without catastrophically forgetting existing knowledge is of critical importance for the development of generally intelligent agents. Most methods devised to address this problem depend heavily on well-defined task boundaries, and thus depend on human supervision. Our task-agnostic method, Self-Activating Neural Ensembles (SANE), uses a modular architecture designed to avoid catastrophic forgetting without making any such assumptions. At the beginning of each trajectory, a module in the SANE ensemble is activated to determine the agent's next policy. During training, new modules are created as needed and only activated modules are updated to ensure that unused modules remain unchanged. This system enables our method to retain and leverage old skills, while growing and learning new ones. We demonstrate our approach on visually rich procedurally generated environments.", "authors": ["Sam Powers", "Eliot Xing", "Abhinav Gupta"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2022-12-31", "url": "https://arxiv.org/abs/2301.00141", "pdf_url": "https://arxiv.org/pdf/2301.00141v1", "arxiv_id": "2301.00141", "doi": "10.48550/arXiv.2301.00141", "citation_count": 9, "influential_citation_count": 1, "has_code": true, "code_url": "https://github.com/AGI-Labs/continual_rl", "venue": "Proceedings of The 1st Conference on Lifelong Learning Agents, PMLR 199:683-704, 2022", "quality_score": 0.25}
{"id": "f6166aef1114fc9df3f6896747571591d8e2e8d03af9d06b2112eb9431ca97a6", "sources": ["arxiv", "semantic_scholar"], "title": "Federated Multi-Agent Deep Reinforcement Learning Approach via Physics-Informed Reward for Multi-Microgrid Energy Management", "abstract": "The utilization of large-scale distributed renewable energy promotes the development of the multi-microgrid (MMG), which raises the need of developing an effective energy management method to minimize economic costs and keep self energy-sufficiency. The multi-agent deep reinforcement learning (MADRL) has been widely used for the energy management problem because of its real-time scheduling ability. However, its training requires massive energy operation data of microgrids (MGs), while gathering these data from different MGs would threaten their privacy and data security. Therefore, this paper tackles this practical yet challenging issue by proposing a federated multi-agent deep reinforcement learning (F-MADRL) algorithm via the physics-informed reward. In this algorithm, the federated learning (FL) mechanism is introduced to train the F-MADRL algorithm thus ensures the privacy and the security of data. In addition, a decentralized MMG model is built, and the energy of each participated MG is managed by an agent, which aims to minimize economic costs and keep self energy-sufficiency according to the physics-informed reward. At first, MGs individually execute the self-training based on local energy operation data to train their local agent models. Then, these local models are periodically uploaded to a server and their parameters are aggregated to build a global agent, which will be broadcasted to MGs and replace their local agents. In this way, the experience of each MG agent can be shared and the energy operation data is not explicitly transmitted, thus protecting the privacy and ensuring data security. Finally, experiments are conducted on Oak Ridge national laboratory distributed energy control communication lab microgrid (ORNL-MG) test system, and the comparisons are carried out to verify the effectiveness of introducing the FL mechanism and the outperformance of our proposed F-MADRL.", "authors": ["Yuanzheng Li", "Shangyang He", "Yang Li", "Yang Shi", "Zhigang Zeng"], "categories": ["eess.SY", "cs.LG"], "fields_of_study": ["Computer Science", "Medicine", "Engineering"], "published_date": "2022-12-29", "url": "https://arxiv.org/abs/2301.00641", "pdf_url": "https://arxiv.org/pdf/2301.00641v1", "arxiv_id": "2301.00641", "doi": "10.1109/TNNLS.2022.3232630", "citation_count": 108, "influential_citation_count": 3, "has_code": false, "code_url": null, "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.5094}
{"id": "c6597b9ad934c2c695914b4b6d39402022831bcf213512333d40dc0dafe37002", "sources": ["arxiv", "semantic_scholar"], "title": "Lifelong Reinforcement Learning with Modulating Masks", "abstract": "Lifelong learning aims to create AI systems that continuously and incrementally learn during a lifetime, similar to biological learning. Attempts so far have met problems, including catastrophic forgetting, interference among tasks, and the inability to exploit previous knowledge. While considerable research has focused on learning multiple supervised classification tasks that involve changes in the input distribution, lifelong reinforcement learning (LRL) must deal with variations in the state and transition distributions, and in the reward functions. Modulating masks with a fixed backbone network, recently developed for classification, are particularly suitable to deal with such a large spectrum of task variations. In this paper, we adapted modulating masks to work with deep LRL, specifically PPO and IMPALA agents. The comparison with LRL baselines in both discrete and continuous RL tasks shows superior performance. We further investigated the use of a linear combination of previously learned masks to exploit previous knowledge when learning new tasks: not only is learning faster, the algorithm solves tasks that we could not otherwise solve from scratch due to extremely sparse rewards. The results suggest that RL with modulating masks is a promising approach to lifelong learning, to the composition of knowledge to learn increasingly complex tasks, and to knowledge reuse for efficient and faster learning.", "authors": ["Eseoghene Ben-Iwhiwhu", "Saptarshi Nath", "Praveen K. Pilly", "Soheil Kolouri", "Andrea Soltoggio"], "categories": ["cs.LG", "cs.AI", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2022-12-21", "url": "https://arxiv.org/abs/2212.11110", "pdf_url": "https://arxiv.org/pdf/2212.11110v3", "arxiv_id": "2212.11110", "doi": "10.48550/arXiv.2212.11110", "citation_count": 29, "influential_citation_count": 3, "has_code": true, "code_url": "https://github.com/dlpbc/mask-lrl", "venue": "Transactions on Machine Learning Research (2023)", "quality_score": 0.3693}
{"id": "fa59d687509005c2db2894131b217409bda5005c5c56c695c42d1a610bde0e9f", "sources": ["arxiv", "semantic_scholar"], "title": "Maximal Initial Learning Rates in Deep ReLU Networks", "abstract": "Training a neural network requires choosing a suitable learning rate, which involves a trade-off between speed and effectiveness of convergence. While there has been considerable theoretical and empirical analysis of how large the learning rate can be, most prior work focuses only on late-stage training. In this work, we introduce the maximal initial learning rate $η^{\\ast}$ - the largest learning rate at which a randomly initialized neural network can successfully begin training and achieve (at least) a given threshold accuracy. Using a simple approach to estimate $η^{\\ast}$, we observe that in constant-width fully-connected ReLU networks, $η^{\\ast}$ behaves differently from the maximum learning rate later in training. Specifically, we find that $η^{\\ast}$ is well predicted as a power of depth $\\times$ width, provided that (i) the width of the network is sufficiently large compared to the depth, and (ii) the input layer is trained at a relatively small learning rate. We further analyze the relationship between $η^{\\ast}$ and the sharpness $λ_{1}$ of the network at initialization, indicating they are closely though not inversely related. We formally prove bounds for $λ_{1}$ in terms of depth $\\times$ width that align with our empirical results.", "authors": ["Gaurav Iyer", "Boris Hanin", "David Rolnick"], "categories": ["stat.ML", "cs.LG"], "fields_of_study": ["Mathematics", "Computer Science"], "published_date": "2022-12-14", "url": "https://arxiv.org/abs/2212.07295", "pdf_url": "https://arxiv.org/pdf/2212.07295v2", "arxiv_id": "2212.07295", "doi": "10.48550/arXiv.2212.07295", "citation_count": 14, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.294}
{"id": "64932d7b2e99af395ea56f4dcb4c90849119bd78a116e4638102b0f3f09a5ae8", "sources": ["arxiv", "semantic_scholar"], "title": "Dual Accuracy-Quality-Driven Neural Network for Prediction Interval Generation", "abstract": "Accurate uncertainty quantification is necessary to enhance the reliability of deep learning models in real-world applications. In the case of regression tasks, prediction intervals (PIs) should be provided along with the deterministic predictions of deep learning models. Such PIs are useful or \"high-quality\" as long as they are sufficiently narrow and capture most of the probability density. In this paper, we present a method to learn prediction intervals for regression-based neural networks automatically in addition to the conventional target predictions. In particular, we train two companion neural networks: one that uses one output, the target estimate, and another that uses two outputs, the upper and lower bounds of the corresponding PI. Our main contribution is the design of a novel loss function for the PI-generation network that takes into account the output of the target-estimation network and has two optimization objectives: minimizing the mean prediction interval width and ensuring the PI integrity using constraints that maximize the prediction interval probability coverage implicitly. Furthermore, we introduce a self-adaptive coefficient that balances both objectives within the loss function, which alleviates the task of fine-tuning. Experiments using a synthetic dataset, eight benchmark datasets, and a real-world crop yield prediction dataset showed that our method was able to maintain a nominal probability coverage and produce significantly narrower PIs without detriment to its target estimation accuracy when compared to those PIs generated by three state-of-the-art neural-network-based methods. In other words, our method was shown to produce higher-quality PIs.", "authors": ["Giorgio Morales", "John W. Sheppard"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics", "Medicine"], "published_date": "2022-12-13", "url": "https://arxiv.org/abs/2212.06370", "pdf_url": "https://arxiv.org/pdf/2212.06370v4", "arxiv_id": "2212.06370", "doi": "10.1109/TNNLS.2023.3339470", "citation_count": 9, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.25}
{"id": "6098ce77e95c6f66be281ac85da8978cabf03fbc34195cbf026ba2cc722214b7", "sources": ["arxiv", "semantic_scholar"], "title": "Overcoming Catastrophic Forgetting by XAI", "abstract": "Explaining the behaviors of deep neural networks, usually considered as black boxes, is critical especially when they are now being adopted over diverse aspects of human life. Taking the advantages of interpretable machine learning (interpretable ML), this work proposes a novel tool called Catastrophic Forgetting Dissector (or CFD) to explain catastrophic forgetting in continual learning settings. We also introduce a new method called Critical Freezing based on the observations of our tool. Experiments on ResNet articulate how catastrophic forgetting happens, particularly showing which components of this famous network are forgetting. Our new continual learning algorithm defeats various recent techniques by a significant margin, proving the capability of the investigation. Critical freezing not only attacks catastrophic forgetting but also exposes explainability.", "authors": ["Giang Nguyen"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2022-11-25", "url": "https://arxiv.org/abs/2211.14177", "pdf_url": "https://arxiv.org/pdf/2211.14177v1", "arxiv_id": "2211.14177", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.0}
{"id": "770bb0ff54a0cb58149d4fd0fa8a52e8266b248d88d19062fc050ec5c56ac3e9", "sources": ["arxiv", "semantic_scholar"], "title": "MECCH: Metapath Context Convolution-based Heterogeneous Graph Neural Networks", "abstract": "Heterogeneous graph neural networks (HGNNs) were proposed for representation learning on structural data with multiple types of nodes and edges. To deal with the performance degradation issue when HGNNs become deep, researchers combine metapaths into HGNNs to associate nodes closely related in semantics but far apart in the graph. However, existing metapath-based models suffer from either information loss or high computation costs. To address these problems, we present a novel Metapath Context Convolution-based Heterogeneous Graph Neural Network (MECCH). MECCH leverages metapath contexts, a new kind of graph structure that facilitates lossless node information aggregation while avoiding any redundancy. Specifically, MECCH applies three novel components after feature preprocessing to extract comprehensive information from the input graph efficiently: (1) metapath context construction, (2) metapath context encoder, and (3) convolutional metapath fusion. Experiments on five real-world heterogeneous graph datasets for node classification and link prediction show that MECCH achieves superior prediction accuracy compared with state-of-the-art baselines with improved computational efficiency.", "authors": ["Xinyu Fu", "Irwin King"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science", "Medicine"], "published_date": "2022-11-23", "url": "https://arxiv.org/abs/2211.12792", "pdf_url": "https://arxiv.org/pdf/2211.12792v2", "arxiv_id": "2211.12792", "doi": "10.1016/j.neunet.2023.11.030", "citation_count": 46, "influential_citation_count": 3, "has_code": true, "code_url": "https://github.com/cynricfu/MECCH", "venue": "Neural Networks", "quality_score": 0.418}
{"id": "744d71073eae2b84cb915a044855d1f67e8e42d6d8d859d2287ce59e4b265b55", "sources": ["arxiv", "semantic_scholar"], "title": "Learning predictive checklists from continuous medical data", "abstract": "Checklists, while being only recently introduced in the medical domain, have become highly popular in daily clinical practice due to their combined effectiveness and great interpretability. Checklists are usually designed by expert clinicians that manually collect and analyze available evidence. However, the increasing quantity of available medical data is calling for a partially automated checklist design. Recent works have taken a step in that direction by learning predictive checklists from categorical data. In this work, we propose to extend this approach to accomodate learning checklists from continuous medical data using mixed-integer programming approach. We show that this extension outperforms a range of explainable machine learning baselines on the prediction of sepsis from intensive care clinical trajectories.", "authors": ["Yukti Makhija", "Edward De Brouwer", "Rahul G. Krishnan"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2022-11-14", "url": "https://arxiv.org/abs/2211.07076", "pdf_url": "https://arxiv.org/pdf/2211.07076v1", "arxiv_id": "2211.07076", "doi": "10.48550/arXiv.2211.07076", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.0753}
{"id": "9a5c0e548277d9bcb6f1e52e71dc8a9fac202de5a9abd4afcd13bccba89a11ea", "sources": ["arxiv", "semantic_scholar"], "title": "Continual Learning by Modeling Intra-Class Variation", "abstract": "It has been observed that neural networks perform poorly when the data or tasks are presented sequentially. Unlike humans, neural networks suffer greatly from catastrophic forgetting, making it impossible to perform life-long learning. To address this issue, memory-based continual learning has been actively studied and stands out as one of the best-performing methods. We examine memory-based continual learning and identify that large variation in the representation space is crucial for avoiding catastrophic forgetting. Motivated by this, we propose to diversify representations by using two types of perturbations: model-agnostic variation (i.e., the variation is generated without the knowledge of the learned neural network) and model-based variation (i.e., the variation is conditioned on the learned neural network). We demonstrate that enlarging representational variation serves as a general principle to improve continual learning. Finally, we perform empirical studies which demonstrate that our method, as a simple plug-and-play component, can consistently improve a number of memory-based continual learning methods by a large margin.", "authors": ["Longhui Yu", "Tianyang Hu", "Lanqing Hong", "Zhen Liu", "Adrian Weller", "Weiyang Liu"], "categories": ["cs.LG", "cs.AI", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2022-10-11", "url": "https://arxiv.org/abs/2210.05398", "pdf_url": "https://arxiv.org/pdf/2210.05398v2", "arxiv_id": "2210.05398", "doi": "10.48550/arXiv.2210.05398", "citation_count": 18, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.3197}
{"id": "7b0c18191dda7d413c47bae95fdcc2eb43c5a73fa35dffb475a4146616f8de8a", "sources": ["arxiv", "semantic_scholar"], "title": "TAME: Task Agnostic Continual Learning using Multiple Experts", "abstract": "The goal of lifelong learning is to continuously learn from non-stationary distributions, where the non-stationarity is typically imposed by a sequence of distinct tasks. Prior works have mostly considered idealistic settings, where the identity of tasks is known at least at training. In this paper we focus on a fundamentally harder, so-called task-agnostic setting where the task identities are not known and the learning machine needs to infer them from the observations. Our algorithm, which we call TAME (Task-Agnostic continual learning using Multiple Experts), automatically detects the shift in data distributions and switches between task expert networks in an online manner. At training, the strategy for switching between tasks hinges on an extremely simple observation that for each new coming task there occurs a statistically-significant deviation in the value of the loss function that marks the onset of this new task. At inference, the switching between experts is governed by the selector network that forwards the test sample to its relevant expert network. The selector network is trained on a small subset of data drawn uniformly at random. We control the growth of the task expert networks as well as selector network by employing online pruning. Our experimental results show the efficacy of our approach on benchmark continual learning data sets, outperforming the previous task-agnostic methods and even the techniques that admit task identities at both training and testing, while at the same time using a comparable model size.", "authors": ["Haoran Zhu", "Maryam Majzoubi", "Arihant Jain", "Anna Choromanska"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2022-10-08", "url": "https://arxiv.org/abs/2210.03869", "pdf_url": "https://arxiv.org/pdf/2210.03869v2", "arxiv_id": "2210.03869", "doi": "10.1109/CVPRW63382.2024.00417", "citation_count": 10, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2603}
{"id": "bc087fe55f82a42c1261b6d3b878da63eaf2f84392a46d3cb31fd1e0cdcb9720", "sources": ["arxiv", "semantic_scholar"], "title": "Kernel Normalized Convolutional Networks for Privacy-Preserving Machine Learning", "abstract": "Normalization is an important but understudied challenge in privacy-related application domains such as federated learning (FL), differential privacy (DP), and differentially private federated learning (DP-FL). While the unsuitability of batch normalization for these domains has already been shown, the impact of other normalization methods on the performance of federated or differentially private models is not well-known. To address this, we draw a performance comparison among layer normalization (LayerNorm), group normalization (GroupNorm), and the recently proposed kernel normalization (KernelNorm) in FL, DP, and DP-FL settings. Our results indicate LayerNorm and GroupNorm provide no performance gain compared to the baseline (i.e. no normalization) for shallow models in FL and DP. They, on the other hand, considerably enhance the performance of shallow models in DP-FL and deeper models in FL and DP. KernelNorm, moreover, significantly outperforms its competitors in terms of accuracy and convergence rate (or communication efficiency) for both shallow and deeper models in all considered learning environments. Given these key observations, we propose a kernel normalized ResNet architecture called KNResNet-13 for differentially private learning. Using the proposed architecture, we provide new state-of-the-art accuracy values on the CIFAR-10 and Imagenette datasets, when trained from scratch.", "authors": ["Reza Nasirigerdeh", "Javad Torkzadehmahani", "Daniel Rueckert", "Georgios Kaissis"], "categories": ["cs.LG", "cs.CR"], "fields_of_study": ["Computer Science"], "published_date": "2022-09-30", "url": "https://arxiv.org/abs/2210.00053", "pdf_url": "https://arxiv.org/pdf/2210.00053v2", "arxiv_id": "2210.00053", "doi": "10.1109/SaTML54575.2023.00016", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "1st IEEE Conference on Secure and Trustworthy Machine Learning (SaTML), 2023", "quality_score": 0.0753}
{"id": "d33ab78c7c2d7f41acba2253ec2ffa51434670964f4179f378847809a4caa623", "sources": ["arxiv", "semantic_scholar"], "title": "Scalable Adversarial Online Continual Learning", "abstract": "Adversarial continual learning is effective for continual learning problems because of the presence of feature alignment process generating task-invariant features having low susceptibility to the catastrophic forgetting problem. Nevertheless, the ACL method imposes considerable complexities because it relies on task-specific networks and discriminators. It also goes through an iterative training process which does not fit for online (one-epoch) continual learning problems. This paper proposes a scalable adversarial continual learning (SCALE) method putting forward a parameter generator transforming common features into task-specific features and a single discriminator in the adversarial game to induce common features. The training process is carried out in meta-learning fashions using a new combination of three loss functions. SCALE outperforms prominent baselines with noticeable margins in both accuracy and execution time.", "authors": ["Tanmoy Dam", "Mahardhika Pratama", "MD Meftahul Ferdaus", "Sreenatha Anavatti", "Hussein Abbas"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2022-09-04", "url": "https://arxiv.org/abs/2209.01558", "pdf_url": "https://arxiv.org/pdf/2209.01558v1", "arxiv_id": "2209.01558", "doi": "10.48550/arXiv.2209.01558", "citation_count": 5, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.1945}
{"id": "36b3cfb9ec6e2df26ac7d598b59138b8f474a7e8c7bdf2528227b85b4bfaf069", "sources": ["arxiv", "semantic_scholar"], "title": "Dynamics-Adaptive Continual Reinforcement Learning via Progressive Contextualization", "abstract": "A key challenge of continual reinforcement learning (CRL) in dynamic environments is to promptly adapt the RL agent's behavior as the environment changes over its lifetime, while minimizing the catastrophic forgetting of the learned information. To address this challenge, in this article, we propose DaCoRL, i.e., dynamics-adaptive continual RL. DaCoRL learns a context-conditioned policy using progressive contextualization, which incrementally clusters a stream of stationary tasks in the dynamic environment into a series of contexts and opts for an expandable multihead neural network to approximate the policy. Specifically, we define a set of tasks with similar dynamics as an environmental context and formalize context inference as a procedure of online Bayesian infinite Gaussian mixture clustering on environment features, resorting to online Bayesian inference to infer the posterior distribution over contexts. Under the assumption of a Chinese restaurant process prior, this technique can accurately classify the current task as a previously seen context or instantiate a new context as needed without relying on any external indicator to signal environmental changes in advance. Furthermore, we employ an expandable multihead neural network whose output layer is synchronously expanded with the newly instantiated context, and a knowledge distillation regularization term for retaining the performance on learned tasks. As a general framework that can be coupled with various deep RL algorithms, DaCoRL features consistent superiority over existing methods in terms of the stability, overall performance and generalization ability, as verified by extensive experiments on several robot navigation and MuJoCo locomotion tasks.", "authors": ["Tiantian Zhang", "Zichuan Lin", "Yuxing Wang", "Deheng Ye", "Qiang Fu", "Wei Yang", "Xueqian Wang", "Bin Liang", "Bo Yuan", "Xiu Li"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science", "Medicine"], "published_date": "2022-09-01", "url": "https://arxiv.org/abs/2209.00347", "pdf_url": "https://arxiv.org/pdf/2209.00347v2", "arxiv_id": "2209.00347", "doi": "10.1109/TNNLS.2023.3280085", "citation_count": 23, "influential_citation_count": 3, "has_code": false, "code_url": null, "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.3451}
{"id": "bdb284f6cca71a8fe39017b87d932713ed96ba3f5555c2e83aca7779ef43d1fb", "sources": ["arxiv", "semantic_scholar"], "title": "A Theory for Knowledge Transfer in Continual Learning", "abstract": "Continual learning of a stream of tasks is an active area in deep neural networks. The main challenge investigated has been the phenomenon of catastrophic forgetting or interference of newly acquired knowledge with knowledge from previous tasks. Recent work has investigated forward knowledge transfer to new tasks. Backward transfer for improving knowledge gained during previous tasks has received much less attention. There is in general limited understanding of how knowledge transfer could aid tasks learned continually. We present a theory for knowledge transfer in continual supervised learning, which considers both forward and backward transfer. We aim at understanding their impact for increasingly knowledgeable learners. We derive error bounds for each of these transfer mechanisms. These bounds are agnostic to specific implementations (e.g. deep neural networks). We demonstrate that, for a continual learner that observes related tasks, both forward and backward transfer can contribute to an increasing performance as more tasks are observed.", "authors": ["Diana Benavides-Prado", "Patricia Riddle"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2022-08-14", "url": "https://arxiv.org/abs/2208.06931", "pdf_url": "https://arxiv.org/pdf/2208.06931v1", "arxiv_id": "2208.06931", "doi": "10.48550/arXiv.2208.06931", "citation_count": 22, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.3404}
{"id": "54a6be676e980490a30924f8b3ceb3fc1e30ee291210d974857595a532aaea47", "sources": ["arxiv", "semantic_scholar"], "title": "Model-Free Generative Replay for Lifelong Reinforcement Learning: Application to Starcraft-2", "abstract": "One approach to meet the challenges of deep lifelong reinforcement learning (LRL) is careful management of the agent's learning experiences, to learn (without forgetting) and build internal meta-models (of the tasks, environments, agents, and world). Generative replay (GR) is a biologically inspired replay mechanism that augments learning experiences with self-labelled examples drawn from an internal generative model that is updated over time. We present a version of GR for LRL that satisfies two desiderata: (a) Introspective density modelling of the latent representations of policies learned using deep RL, and (b) Model-free end-to-end learning. In this paper, we study three deep learning architectures for model-free GR, starting from a naïve GR and adding ingredients to achieve (a) and (b). We evaluate our proposed algorithms on three different scenarios comprising tasks from the Starcraft-2 and Minigrid domains. We report several key findings showing the impact of the design choices on quantitative metrics that include transfer learning, generalization to unseen tasks, fast adaptation after task change, performance wrt task expert, and catastrophic forgetting. We observe that our GR prevents drift in the features-to-action mapping from the latent vector space of a deep RL agent. We also show improvements in established lifelong learning metrics. We find that a small random replay buffer significantly increases the stability of training. Overall, we find that \"hidden replay\" (a well-known architecture for class-incremental classification) is the most promising approach that pushes the state-of-the-art in GR for LRL and observe that the architecture of the sleep model might be more important for improving performance than the types of replay used. Our experiments required only 6% of training samples to achieve 80-90% of expert performance in most Starcraft-2 scenarios.", "authors": ["Zachary Daniels", "Aswin Raghavan", "Jesse Hostetler", "Abrar Rahman", "Indranil Sur", "Michael Piacentino", "Ajay Divakaran"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2022-08-09", "url": "https://arxiv.org/abs/2208.05056", "pdf_url": "https://arxiv.org/pdf/2208.05056v2", "arxiv_id": "2208.05056", "doi": "10.48550/arXiv.2208.05056", "citation_count": 16, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.3076}
{"id": "3cdc1cfc55251e4e66ed77cb2cb13db28e939805fa4c318c69e760b7e94c2928", "sources": ["arxiv", "semantic_scholar"], "title": "Uncertain Bayesian Networks: Learning from Incomplete Data", "abstract": "When the historical data are limited, the conditional probabilities associated with the nodes of Bayesian networks are uncertain and can be empirically estimated. Second order estimation methods provide a framework for both estimating the probabilities and quantifying the uncertainty in these estimates. We refer to these cases as uncer tain or second-order Bayesian networks. When such data are complete, i.e., all variable values are observed for each instantiation, the conditional probabilities are known to be Dirichlet-distributed. This paper improves the current state-of-the-art approaches for handling uncertain Bayesian networks by enabling them to learn distributions for their parameters, i.e., conditional probabilities, with incomplete data. We extensively evaluate various methods to learn the posterior of the parameters through the desired and empirically derived strength of confidence bounds for various queries.", "authors": ["Conrad D. Hougen", "Lance M. Kaplan", "Federico Cerutti", "Alfred O. Hero"], "categories": ["stat.ML", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2022-08-08", "url": "https://arxiv.org/abs/2208.04221", "pdf_url": "https://arxiv.org/pdf/2208.04221v1", "arxiv_id": "2208.04221", "doi": "10.1109/MLSP52302.2021.9596205", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "International Workshop on Machine Learning for Signal Processing", "quality_score": 0.0753}
{"id": "501baf15bdceb66dbb74585868cf97bb989a297f171eb2abd156ab2657651790", "sources": ["arxiv", "semantic_scholar"], "title": "Centroids Matching: an efficient Continual Learning approach operating in the embedding space", "abstract": "Catastrophic forgetting (CF) occurs when a neural network loses the information previously learned while training on a set of samples from a different distribution, i.e., a new task. Existing approaches have achieved remarkable results in mitigating CF, especially in a scenario called task incremental learning. However, this scenario is not realistic, and limited work has been done to achieve good results on more realistic scenarios. In this paper, we propose a novel regularization method called Centroids Matching, that, inspired by meta-learning approaches, fights CF by operating in the feature space produced by the neural network, achieving good results while requiring a small memory footprint. Specifically, the approach classifies the samples directly using the feature vectors produced by the neural network, by matching those vectors with the centroids representing the classes from the current task, or all the tasks up to that point. Centroids Matching is faster than competing baselines, and it can be exploited to efficiently mitigate CF, by preserving the distances between the embedding space produced by the model when past tasks were over, and the one currently produced, leading to a method that achieves high accuracy on all the tasks, without using an external memory when operating on easy scenarios, or using a small one for more realistic ones. Extensive experiments demonstrate that Centroids Matching achieves accuracy gains on multiple datasets and scenarios.", "authors": ["Jary Pomponi", "Simone Scardapane", "Aurelio Uncini"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2022-08-03", "url": "https://arxiv.org/abs/2208.02048", "pdf_url": "https://arxiv.org/pdf/2208.02048v2", "arxiv_id": "2208.02048", "doi": "10.48550/arXiv.2208.02048", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.0753}
{"id": "24fcc9c779bfd8d54c89e2df45490fb283645995e4119aa2312a0df6ba6d5574", "sources": ["arxiv", "semantic_scholar"], "title": "Latent Properties of Lifelong Learning Systems", "abstract": "Creating artificial intelligence (AI) systems capable of demonstrating lifelong learning is a fundamental challenge, and many approaches and metrics have been proposed to analyze algorithmic properties. However, for existing lifelong learning metrics, algorithmic contributions are confounded by task and scenario structure. To mitigate this issue, we introduce an algorithm-agnostic explainable surrogate-modeling approach to estimate latent properties of lifelong learning algorithms. We validate the approach for estimating these properties via experiments on synthetic data. To validate the structure of the surrogate model, we analyze real performance data from a collection of popular lifelong learning approaches and baselines adapted for lifelong classification and lifelong reinforcement learning.", "authors": ["Corban Rivera", "Chace Ashcraft", "Alexander New", "James Schmidt", "Gautam Vallabha"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2022-07-28", "url": "https://arxiv.org/abs/2207.14378", "pdf_url": "https://arxiv.org/pdf/2207.14378v1", "arxiv_id": "2207.14378", "doi": "10.48550/arXiv.2207.14378", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.0753}
{"id": "27c31e4162d5754f1edf41e8322ae0cde0e887ebbcc779cb0a0edaf6af758c15", "sources": ["arxiv", "semantic_scholar"], "title": "Federated Learning and catastrophic forgetting in pervasive computing: demonstration in HAR domain", "abstract": "Federated Learning has been introduced as a new machine learning paradigm enhancing the use of local devices. At a server level, FL regularly aggregates models learned locally on distributed clients to obtain a more general model. In this way, no private data is sent over the network, and the communication cost is reduced. However, current solutions rely on the availability of large amounts of stored data at the client side in order to fine-tune the models sent by the server. Such setting is not realistic in mobile pervasive computing where data storage must be kept low and data characteristic (distribution) can change dramatically. To account for this variability, a solution is to use the data regularly collected by the client to progressively adapt the received model. But such naive approach exposes clients to the well-known problem of catastrophic forgetting. The purpose of this paper is to demonstrate this problem in the mobile human activity recognition context on smartphones.", "authors": ["Anastasiia Usmanova", "François Portet", "Philippe Lalanda", "German Vega"], "categories": ["cs.LG", "cs.AI", "eess.SP"], "fields_of_study": ["Computer Science", "Engineering"], "published_date": "2022-07-17", "url": "https://arxiv.org/abs/2207.08180", "pdf_url": "https://arxiv.org/pdf/2207.08180v1", "arxiv_id": "2207.08180", "doi": "10.1109/PerComWorkshops53856.2022.9767246", "citation_count": 4, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.1747}
{"id": "a1ecc5222a4e05d98931d59c416473d5810eb056746ada5bb1d0b62dea709202", "sources": ["arxiv", "semantic_scholar"], "title": "Federated Continual Learning through distillation in pervasive computing", "abstract": "Federated Learning has been introduced as a new machine learning paradigm enhancing the use of local devices. At a server level, FL regularly aggregates models learned locally on distributed clients to obtain a more general model. Current solutions rely on the availability of large amounts of stored data at the client side in order to fine-tune the models sent by the server. Such setting is not realistic in mobile pervasive computing where data storage must be kept low and data characteristic can change dramatically. To account for this variability, a solution is to use the data regularly collected by the client to progressively adapt the received model. But such naive approach exposes clients to the well-known problem of catastrophic forgetting. To address this problem, we have defined a Federated Continual Learning approach which is mainly based on distillation. Our approach allows a better use of resources, eliminating the need to retrain from scratch at the arrival of new data and reducing memory usage by limiting the amount of data to be stored. This proposal has been evaluated in the Human Activity Recognition (HAR) domain and has shown to effectively reduce the catastrophic forgetting effect.", "authors": ["Anastasiia Usmanova", "François Portet", "Philippe Lalanda", "German Vega"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2022-07-17", "url": "https://arxiv.org/abs/2207.08181", "pdf_url": "https://arxiv.org/pdf/2207.08181v1", "arxiv_id": "2207.08181", "doi": "10.1109/smartcomp55677.2022.00027", "citation_count": 15, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "International Conference on Smart Computing", "quality_score": 0.301}
{"id": "31ddaa1bc87bb291f6e54a054dfc1ecd34ebebac77847e66fcc04a4213ab482a", "sources": ["arxiv", "semantic_scholar"], "title": "Task Agnostic Representation Consolidation: a Self-supervised based Continual Learning Approach", "abstract": "Continual learning (CL) over non-stationary data streams remains one of the long-standing challenges in deep neural networks (DNNs) as they are prone to catastrophic forgetting. CL models can benefit from self-supervised pre-training as it enables learning more generalizable task-agnostic features. However, the effect of self-supervised pre-training diminishes as the length of task sequences increases. Furthermore, the domain shift between pre-training data distribution and the task distribution reduces the generalizability of the learned representations. To address these limitations, we propose Task Agnostic Representation Consolidation (TARC), a two-stage training paradigm for CL that intertwines task-agnostic and task-specific learning whereby self-supervised training is followed by supervised learning for each task. To further restrict the deviation from the learned representations in the self-supervised stage, we employ a task-agnostic auxiliary loss during the supervised stage. We show that our training paradigm can be easily added to memory- or regularization-based approaches and provides consistent performance gain across more challenging CL settings. We further show that it leads to more robust and well-calibrated models.", "authors": ["Prashant Bhat", "Bahram Zonooz", "Elahe Arani"], "categories": ["cs.LG", "cs.AI", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2022-07-13", "url": "https://arxiv.org/abs/2207.06267", "pdf_url": "https://arxiv.org/pdf/2207.06267v1", "arxiv_id": "2207.06267", "doi": "10.48550/arXiv.2207.06267", "citation_count": 15, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.301}
{"id": "45f3b9fcac372d02e9a313d5193c35177cec749c9ddbdf27a8e3e570779a1ba2", "sources": ["arxiv", "semantic_scholar"], "title": "Consistency is the key to further mitigating catastrophic forgetting in continual learning", "abstract": "Deep neural networks struggle to continually learn multiple sequential tasks due to catastrophic forgetting of previously learned tasks. Rehearsal-based methods which explicitly store previous task samples in the buffer and interleave them with the current task samples have proven to be the most effective in mitigating forgetting. However, Experience Replay (ER) does not perform well under low-buffer regimes and longer task sequences as its performance is commensurate with the buffer size. Consistency in predictions of soft-targets can assist ER in preserving information pertaining to previous tasks better as soft-targets capture the rich similarity structure of the data. Therefore, we examine the role of consistency regularization in ER framework under various continual learning scenarios. We also propose to cast consistency regularization as a self-supervised pretext task thereby enabling the use of a wide variety of self-supervised learning methods as regularizers. While simultaneously enhancing model calibration and robustness to natural corruptions, regularizing consistency in predictions results in lesser forgetting across all continual learning scenarios. Among the different families of regularizers, we find that stricter consistency constraints preserve previous task information in ER better.", "authors": ["Prashant Bhat", "Bahram Zonooz", "Elahe Arani"], "categories": ["cs.LG", "cs.AI", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2022-07-11", "url": "https://arxiv.org/abs/2207.04998", "pdf_url": "https://arxiv.org/pdf/2207.04998v1", "arxiv_id": "2207.04998", "doi": "10.48550/arXiv.2207.04998", "citation_count": 19, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.3253}
{"id": "9103ad1c6a4eb2568e02b1cee55ceb64b0245a37244ffbb175420ec5accc1fd1", "sources": ["arxiv", "semantic_scholar"], "title": "An Introduction to Lifelong Supervised Learning", "abstract": "This primer is an attempt to provide a detailed summary of the different facets of lifelong learning. We start with Chapter 2 which provides a high-level overview of lifelong learning systems. In this chapter, we discuss prominent scenarios in lifelong learning (Section 2.4), provide 8 Introduction a high-level organization of different lifelong learning approaches (Section 2.5), enumerate the desiderata for an ideal lifelong learning system (Section 2.6), discuss how lifelong learning is related to other learning paradigms (Section 2.7), describe common metrics used to evaluate lifelong learning systems (Section 2.8). This chapter is more useful for readers who are new to lifelong learning and want to get introduced to the field without focusing on specific approaches or benchmarks. The remaining chapters focus on specific aspects (either learning algorithms or benchmarks) and are more useful for readers who are looking for specific approaches or benchmarks. Chapter 3 focuses on regularization-based approaches that do not assume access to any data from previous tasks. Chapter 4 discusses memory-based approaches that typically use a replay buffer or an episodic memory to save subset of data across different tasks. Chapter 5 focuses on different architecture families (and their instantiations) that have been proposed for training lifelong learning systems. Following these different classes of learning algorithms, we discuss the commonly used evaluation benchmarks and metrics for lifelong learning (Chapter 6) and wrap up with a discussion of future challenges and important research directions in Chapter 7.", "authors": ["Shagun Sodhani", "Mojtaba Faramarzi", "Sanket Vaibhav Mehta", "Pranshu Malviya", "Mohamed Abdelsalam", "Janarthanan Janarthanan", "Sarath Chandar"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2022-07-10", "url": "https://arxiv.org/abs/2207.04354", "pdf_url": "https://arxiv.org/pdf/2207.04354v2", "arxiv_id": "2207.04354", "doi": "10.48550/arXiv.2207.04354", "citation_count": 23, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3451}
{"id": "b7e4e6d0738b28edbfe1247ebf256820e1f04dfe5366e247af944842494bb64d", "sources": ["arxiv", "semantic_scholar"], "title": "Federated and Transfer Learning: A Survey on Adversaries and Defense Mechanisms", "abstract": "The advent of federated learning has facilitated large-scale data exchange amongst machine learning models while maintaining privacy. Despite its brief history, federated learning is rapidly evolving to make wider use more practical. One of the most significant advancements in this domain is the incorporation of transfer learning into federated learning, which overcomes fundamental constraints of primary federated learning, particularly in terms of security. This chapter performs a comprehensive survey on the intersection of federated and transfer learning from a security point of view. The main goal of this study is to uncover potential vulnerabilities and defense mechanisms that might compromise the privacy and performance of systems that use federated and transfer learning.", "authors": ["Ehsan Hallaji", "Roozbeh Razavi-Far", "Mehrdad Saif"], "categories": ["cs.LG", "cs.AI", "cs.CR", "cs.CV", "cs.DC"], "fields_of_study": ["Computer Science"], "published_date": "2022-07-05", "url": "https://arxiv.org/abs/2207.02337", "pdf_url": "https://arxiv.org/pdf/2207.02337v1", "arxiv_id": "2207.02337", "doi": "10.1007/978-3-031-11748-0_3", "citation_count": 16, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3076}
{"id": "b5a99a3782bd48051a2733ef8df0b811027b77e960acfc74eb29711eb6c6a56c", "sources": ["arxiv", "semantic_scholar"], "title": "Modular Lifelong Reinforcement Learning via Neural Composition", "abstract": "Humans commonly solve complex problems by decomposing them into easier subproblems and then combining the subproblem solutions. This type of compositional reasoning permits reuse of the subproblem solutions when tackling future tasks that share part of the underlying compositional structure. In a continual or lifelong reinforcement learning (RL) setting, this ability to decompose knowledge into reusable components would enable agents to quickly learn new RL tasks by leveraging accumulated compositional structures. We explore a particular form of composition based on neural modules and present a set of RL problems that intuitively admit compositional solutions. Empirically, we demonstrate that neural composition indeed captures the underlying structure of this space of problems. We further propose a compositional lifelong RL method that leverages accumulated neural components to accelerate the learning of future tasks while retaining performance on previous tasks via off-line RL over replayed experiences.", "authors": ["Jorge A. Mendez", "Harm van Seijen", "Eric Eaton"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2022-07-01", "url": "https://arxiv.org/abs/2207.00429", "pdf_url": "https://arxiv.org/pdf/2207.00429v1", "arxiv_id": "2207.00429", "doi": "10.48550/arXiv.2207.00429", "citation_count": 51, "influential_citation_count": 1, "has_code": true, "code_url": "https://github.com/Lifelong-ML/Mendez2022ModularLifelongRL", "venue": "International Conference on Learning Representations", "quality_score": 0.429}
{"id": "cf6c12cdddeead3de365ed916a0478a0d1b9ef7e0a8881ba482a727c5c908d0f", "sources": ["arxiv", "semantic_scholar"], "title": "Lifelong Inverse Reinforcement Learning", "abstract": "Methods for learning from demonstration (LfD) have shown success in acquiring behavior policies by imitating a user. However, even for a single task, LfD may require numerous demonstrations. For versatile agents that must learn many tasks via demonstration, this process would substantially burden the user if each task were learned in isolation. To address this challenge, we introduce the novel problem of lifelong learning from demonstration, which allows the agent to continually build upon knowledge learned from previously demonstrated tasks to accelerate the learning of new tasks, reducing the amount of demonstrations required. As one solution to this problem, we propose the first lifelong learning approach to inverse reinforcement learning, which learns consecutive tasks via demonstration, continually transferring knowledge between tasks to improve performance.", "authors": ["Jorge A. Mendez", "Shashank Shivkumar", "Eric Eaton"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2022-07-01", "url": "https://arxiv.org/abs/2207.00461", "pdf_url": "https://arxiv.org/pdf/2207.00461v1", "arxiv_id": "2207.00461", "doi": "10.48550/arXiv.2207.00461", "citation_count": 23, "influential_citation_count": 1, "has_code": true, "code_url": "https://github.com/Lifelong-ML/ELIRL", "venue": "Neural Information Processing Systems", "quality_score": 0.3451}
{"id": "c862b60dcddf4fa2cf336bf4760223eb30e57388e42d259bc3ca7ad487757bf8", "sources": ["arxiv", "semantic_scholar"], "title": "NISPA: Neuro-Inspired Stability-Plasticity Adaptation for Continual Learning in Sparse Networks", "abstract": "The goal of continual learning (CL) is to learn different tasks over time. The main desiderata associated with CL are to maintain performance on older tasks, leverage the latter to improve learning of future tasks, and to introduce minimal overhead in the training process (for instance, to not require a growing model or retraining). We propose the Neuro-Inspired Stability-Plasticity Adaptation (NISPA) architecture that addresses these desiderata through a sparse neural network with fixed density. NISPA forms stable paths to preserve learned knowledge from older tasks. Also, NISPA uses connection rewiring to create new plastic paths that reuse existing knowledge on novel tasks. Our extensive evaluation on EMNIST, FashionMNIST, CIFAR10, and CIFAR100 datasets shows that NISPA significantly outperforms representative state-of-the-art continual learning baselines, and it uses up to ten times fewer learnable parameters compared to baselines. We also make the case that sparsity is an essential ingredient for continual learning. The NISPA code is available at https://github.com/BurakGurbuz97/NISPA.", "authors": ["Mustafa Burak Gurbuz", "Constantine Dovrolis"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2022-06-18", "url": "https://arxiv.org/abs/2206.09117", "pdf_url": "https://arxiv.org/pdf/2206.09117v1", "arxiv_id": "2206.09117", "doi": "10.48550/arXiv.2206.09117", "citation_count": 56, "influential_citation_count": 6, "has_code": true, "code_url": "https://github.com/BurakGurbuz97/NISPA", "venue": "International Conference on Machine Learning", "quality_score": 0.439}
{"id": "465933d948c88afcf144da6787f6300aca815629b6666f4fcb097ba84743e14c", "sources": ["arxiv", "semantic_scholar"], "title": "Lifelong Wandering: A realistic few-shot online continual learning setting", "abstract": "Online few-shot learning describes a setting where models are trained and evaluated on a stream of data while learning emerging classes. While prior work in this setting has achieved very promising performance on instance classification when learning from data-streams composed of a single indoor environment, we propose to extend this setting to consider object classification on a series of several indoor environments, which is likely to occur in applications such as robotics. Importantly, our setting, which we refer to as online few-shot continual learning, injects the well-studied issue of catastrophic forgetting into the few-shot online learning paradigm. In this work, we benchmark several existing methods and adapted baselines within our setting, and show there exists a trade-off between catastrophic forgetting and online performance. Our findings motivate the need for future work in this setting, which can achieve better online performance without catastrophic forgetting.", "authors": ["Mayank Lunayach", "James Smith", "Zsolt Kira"], "categories": ["cs.CV", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2022-06-16", "url": "https://arxiv.org/abs/2206.07932", "pdf_url": "https://arxiv.org/pdf/2206.07932v1", "arxiv_id": "2206.07932", "doi": "10.48550/arXiv.2206.07932", "citation_count": 3, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1505}
{"id": "5103ca9b57e06127ce91a3dd2fe68c8a14c8076681546a69e86182e49f23d396", "sources": ["arxiv", "semantic_scholar"], "title": "Few-Shot Learning by Dimensionality Reduction in Gradient Space", "abstract": "We introduce SubGD, a novel few-shot learning method which is based on the recent finding that stochastic gradient descent updates tend to live in a low-dimensional parameter subspace. In experimental and theoretical analyses, we show that models confined to a suitable predefined subspace generalize well for few-shot learning. A suitable subspace fulfills three criteria across the given tasks: it (a) allows to reduce the training error by gradient flow, (b) leads to models that generalize well, and (c) can be identified by stochastic gradient descent. SubGD identifies these subspaces from an eigendecomposition of the auto-correlation matrix of update directions across different tasks. Demonstrably, we can identify low-dimensional suitable subspaces for few-shot learning of dynamical systems, which have varying properties described by one or few parameters of the analytical system description. Such systems are ubiquitous among real-world applications in science and engineering. We experimentally corroborate the advantages of SubGD on three distinct dynamical systems problem settings, significantly outperforming popular few-shot learning methods both in terms of sample efficiency and performance.", "authors": ["Martin Gauch", "Maximilian Beck", "Thomas Adler", "Dmytro Kotsur", "Stefan Fiel", "Hamid Eghbal-zadeh", "Johannes Brandstetter", "Johannes Kofler", "Markus Holzleitner", "Werner Zellinger", "Daniel Klotz", "Sepp Hochreiter", "Sebastian Lehner"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2022-06-07", "url": "https://arxiv.org/abs/2206.03483", "pdf_url": "https://arxiv.org/pdf/2206.03483v1", "arxiv_id": "2206.03483", "doi": "10.48550/arXiv.2206.03483", "citation_count": 11, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/ml-jku/subgd", "venue": "Proceedings of The 1st Conference on Lifelong Learning Agents, PMLR 199:1043-1064 (2022)", "quality_score": 0.2698}
{"id": "1693803ddf3539c934dfdbf489be968a9eac8cf4e8e7d8b321bfcefbe3068c2e", "sources": ["arxiv", "semantic_scholar"], "title": "Discretization Invariant Networks for Learning Maps between Neural Fields", "abstract": "With the emergence of powerful representations of continuous data in the form of neural fields, there is a need for discretization invariant learning: an approach for learning maps between functions on continuous domains without being sensitive to how the function is sampled. We present a new framework for understanding and designing discretization invariant neural networks (DI-Nets), which generalizes many discrete networks such as convolutional neural networks as well as continuous networks such as neural operators. Our analysis establishes upper bounds on the deviation in model outputs under different finite discretizations, and highlights the central role of point set discrepancy in characterizing such bounds. This insight leads to the design of a family of neural networks driven by numerical integration via quasi-Monte Carlo sampling with discretizations of low discrepancy. We prove by construction that DI-Nets universally approximate a large class of maps between integrable function spaces, and show that discretization invariance also describes backpropagation through such models. Applied to neural fields, convolutional DI-Nets can learn to classify and segment visual data under various discretizations, and sometimes generalize to new types of discretizations at test time. Code: https://github.com/clintonjwang/DI-net.", "authors": ["Clinton J. Wang", "Polina Golland"], "categories": ["cs.LG", "cs.CV", "cs.NE"], "fields_of_study": ["Computer Science"], "published_date": "2022-06-02", "url": "https://arxiv.org/abs/2206.01178", "pdf_url": "https://arxiv.org/pdf/2206.01178v4", "arxiv_id": "2206.01178", "doi": null, "citation_count": 1, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/clintonjwang/DI-net", "venue": null, "quality_score": 0.0753}
{"id": "3f844961f61eb5884e96708f8bdea24cecc0005da458b619075b9da2b9cb4475", "sources": ["arxiv", "semantic_scholar"], "title": "Feature Forgetting in Continual Representation Learning", "abstract": "In continual and lifelong learning, good representation learning can help increase performance and reduce sample complexity when learning new tasks. There is evidence that representations do not suffer from \"catastrophic forgetting\" even in plain continual learning, but little further fact is known about its characteristics. In this paper, we aim to gain more understanding about representation learning in continual learning, especially on the feature forgetting problem. We devise a protocol for evaluating representation in continual learning, and then use it to present an overview of the basic trends of continual representation learning, showing its consistent deficiency and potential issues. To study the feature forgetting problem, we create a synthetic dataset to identify and visualize the prevalence of feature forgetting in neural networks. Finally, we propose a simple technique using gating adapters to mitigate feature forgetting. We conclude by discussing that improving representation learning benefits both old and new tasks in continual learning.", "authors": ["Xiao Zhang", "Dejing Dou", "Ji Wu"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2022-05-26", "url": "https://arxiv.org/abs/2205.13359", "pdf_url": "https://arxiv.org/pdf/2205.13359v1", "arxiv_id": "2205.13359", "doi": "10.48550/arXiv.2205.13359", "citation_count": 8, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2386}
{"id": "a03a2d55c1d3d493605b906d00846c8b1773d0489bda6d34c01ef0471a61f9fb", "sources": ["arxiv", "semantic_scholar"], "title": "Acute Lymphoblastic Leukemia Detection Using Hypercomplex-Valued Convolutional Neural Networks", "abstract": "This paper features convolutional neural networks defined on hypercomplex algebras applied to classify lymphocytes in blood smear digital microscopic images. Such classification is helpful for the diagnosis of acute lymphoblast leukemia (ALL), a type of blood cancer. We perform the classification task using eight hypercomplex-valued convolutional neural networks (HvCNNs) along with real-valued convolutional networks. Our results show that HvCNNs perform better than the real-valued model, showcasing higher accuracy with a much smaller number of parameters. Moreover, we found that HvCNNs based on Clifford algebras processing HSV-encoded images attained the highest observed accuracies. Precisely, our HvCNN yielded an average accuracy rate of 96.6% using the ALL-IDB2 dataset with a 50% train-test split, a value extremely close to the state-of-the-art models but using a much simpler architecture with significantly fewer parameters.", "authors": ["Guilherme Vieira", "Marcos Eduardo Valle"], "categories": ["cs.CV", "cs.LG", "cs.NE", "eess.IV"], "fields_of_study": ["Computer Science", "Engineering"], "published_date": "2022-05-26", "url": "https://arxiv.org/abs/2205.13273", "pdf_url": "https://arxiv.org/pdf/2205.13273v1", "arxiv_id": "2205.13273", "doi": "10.1109/IJCNN55064.2022.9892036", "citation_count": 26, "influential_citation_count": 5, "has_code": false, "code_url": null, "venue": "IEEE International Joint Conference on Neural Network", "quality_score": 0.3891}
{"id": "349c5ca89598f57ffbbdad0e0bc913d001d9c9bc5d8018ef446cfc28ca8e55ad", "sources": ["arxiv", "semantic_scholar"], "title": "Explanatory machine learning for sequential human teaching", "abstract": "The topic of comprehensibility of machine-learned theories has recently drawn increasing attention. Inductive Logic Programming (ILP) uses logic programming to derive logic theories from small data based on abduction and induction techniques. Learned theories are represented in the form of rules as declarative descriptions of obtained knowledge. In earlier work, the authors provided the first evidence of a measurable increase in human comprehension based on machine-learned logic rules for simple classification tasks. In a later study, it was found that the presentation of machine-learned explanations to humans can produce both beneficial and harmful effects in the context of game learning. We continue our investigation of comprehensibility by examining the effects of the ordering of concept presentations on human comprehension. In this work, we examine the explanatory effects of curriculum order and the presence of machine-learned explanations for sequential problem-solving. We show that 1) there exist tasks A and B such that learning A before B has a better human comprehension with respect to learning B before A and 2) there exist tasks A and B such that the presence of explanations when learning A contributes to improved human comprehension when subsequently learning B. We propose a framework for the effects of sequential teaching on comprehension based on an existing definition of comprehensibility and provide evidence for support from data collected in human trials. Empirical results show that sequential teaching of concepts with increasing complexity a) has a beneficial effect on human comprehension and b) leads to human re-discovery of divide-and-conquer problem-solving strategies, and c) studying machine-learned explanations allows adaptations of human problem-solving strategy with better performance.", "authors": ["Lun Ai", "Johannes Langer", "Stephen H. Muggleton", "Ute Schmid"], "categories": ["cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2022-05-20", "url": "https://arxiv.org/abs/2205.10250", "pdf_url": "https://arxiv.org/pdf/2205.10250v2", "arxiv_id": "2205.10250", "doi": "10.1007/s10994-023-06351-8", "citation_count": 8, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Machine-mediated learning", "quality_score": 0.2386}
{"id": "85c531ddee786e57de0f8e67051354ef27f1d5721cad6230f39cc0d1abab7aff", "sources": ["arxiv", "semantic_scholar"], "title": "Transfer Learning with Pre-trained Conditional Generative Models", "abstract": "Transfer learning is crucial in training deep neural networks on new target tasks. Current transfer learning methods always assume at least one of (i) source and target task label spaces overlap, (ii) source datasets are available, and (iii) target network architectures are consistent with source ones. However, holding these assumptions is difficult in practical settings because the target task rarely has the same labels as the source task, the source dataset access is restricted due to storage costs and privacy, and the target architecture is often specialized to each task. To transfer source knowledge without these assumptions, we propose a transfer learning method that uses deep generative models and is composed of the following two stages: pseudo pre-training (PP) and pseudo semi-supervised learning (P-SSL). PP trains a target architecture with an artificial dataset synthesized by using conditional source generative models. P-SSL applies SSL algorithms to labeled target data and unlabeled pseudo samples, which are generated by cascading the source classifier and generative models to condition them with target samples. Our experimental results indicate that our method can outperform the baselines of scratch training and knowledge distillation.", "authors": ["Shin'ya Yamaguchi", "Sekitoshi Kanai", "Atsutoshi Kumagai", "Daiki Chijiwa", "Hisashi Kashima"], "categories": ["cs.LG", "cs.AI", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2022-04-27", "url": "https://arxiv.org/abs/2204.12833", "pdf_url": "https://arxiv.org/pdf/2204.12833v3", "arxiv_id": "2204.12833", "doi": "10.1007/s10994-025-06748-7", "citation_count": 7, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "Machine-mediated learning", "quality_score": 0.2258}
{"id": "ddcc258df45fb681b4bbd468d7cdeb29cda3f2e4d6083fb945a177f3a1eecb81", "sources": ["arxiv", "semantic_scholar"], "title": "On Feature Learning in Neural Networks with Global Convergence Guarantees", "abstract": "We study the optimization of wide neural networks (NNs) via gradient flow (GF) in setups that allow feature learning while admitting non-asymptotic global convergence guarantees. First, for wide shallow NNs under the mean-field scaling and with a general class of activation functions, we prove that when the input dimension is no less than the size of the training set, the training loss converges to zero at a linear rate under GF. Building upon this analysis, we study a model of wide multi-layer NNs whose second-to-last layer is trained via GF, for which we also prove a linear-rate convergence of the training loss to zero, but regardless of the input dimension. We also show empirically that, unlike in the Neural Tangent Kernel (NTK) regime, our multi-layer model exhibits feature learning and can achieve better generalization performance than its NTK counterpart.", "authors": ["Zhengdao Chen", "Eric Vanden-Eijnden", "Joan Bruna"], "categories": ["cs.LG", "math.OC", "math.PR", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2022-04-22", "url": "https://arxiv.org/abs/2204.10782", "pdf_url": "https://arxiv.org/pdf/2204.10782v1", "arxiv_id": "2204.10782", "doi": "10.48550/arXiv.2204.10782", "citation_count": 15, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.301}
{"id": "cd3ed73e79a90f92d1b8dd064d0b64dc3fe530422e8105b3767cd10a8e3e88ea", "sources": ["arxiv", "semantic_scholar"], "title": "CNLL: A Semi-supervised Approach For Continual Noisy Label Learning", "abstract": "The task of continual learning requires careful design of algorithms that can tackle catastrophic forgetting. However, the noisy label, which is inevitable in a real-world scenario, seems to exacerbate the situation. While very few studies have addressed the issue of continual learning under noisy labels, long training time and complicated training schemes limit their applications in most cases. In contrast, we propose a simple purification technique to effectively cleanse the online data stream that is both cost-effective and more accurate. After purification, we perform fine-tuning in a semi-supervised fashion that ensures the participation of all available samples. Training in this fashion helps us learn a better representation that results in state-of-the-art (SOTA) performance. Through extensive experimentation on 3 benchmark datasets, MNIST, CIFAR10 and CIFAR100, we show the effectiveness of our proposed approach. We achieve a 24.8% performance gain for CIFAR10 with 20% noise over previous SOTA methods. Our code is publicly available.", "authors": ["Nazmul Karim", "Umar Khalid", "Ashkan Esmaeili", "Nazanin Rahnavard"], "categories": ["cs.CV", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2022-04-21", "url": "https://arxiv.org/abs/2204.09881", "pdf_url": "https://arxiv.org/pdf/2204.09881v1", "arxiv_id": "2204.09881", "doi": "10.1109/CVPRW56347.2022.00433", "citation_count": 25, "influential_citation_count": 5, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.3891}
{"id": "3444e2077930eb72226d9a13ed45fc515434e2d28a19458cd4e2abb0209d8c06", "sources": ["arxiv", "semantic_scholar"], "title": "Forgetting and Imbalance in Robot Lifelong Learning with Off-policy Data", "abstract": "Robots will experience non-stationary environment dynamics throughout their lifetime: the robot dynamics can change due to wear and tear, or its surroundings may change over time. Eventually, the robots should perform well in all of the environment variations it has encountered. At the same time, it should still be able to learn fast in a new environment. We identify two challenges in Reinforcement Learning (RL) under such a lifelong learning setting with off-policy data: first, existing off-policy algorithms struggle with the trade-off between being conservative to maintain good performance in the old environment and learning efficiently in the new environment, despite keeping all the data in the replay buffer. We propose the Offline Distillation Pipeline to break this trade-off by separating the training procedure into an online interaction phase and an offline distillation phase.Second, we find that training with the imbalanced off-policy data from multiple environments across the lifetime creates a significant performance drop. We identify that this performance drop is caused by the combination of the imbalanced quality and size among the datasets which exacerbate the extrapolation error of the Q-function. During the distillation phase, we apply a simple fix to the issue by keeping the policy closer to the behavior policy that generated the data. In the experiments, we demonstrate these two challenges and the proposed solutions with a simulated bipedal robot walk-ing task across various environment changes. We show that the Offline Distillation Pipeline achieves better performance across all the encountered environments without affecting data collection. We also provide a comprehensive empirical study to support our hypothesis on the data imbalance issue.", "authors": ["Wenxuan Zhou", "Steven Bohez", "Jan Humplik", "Abbas Abdolmaleki", "Dushyant Rao", "Markus Wulfmeier", "Tuomas Haarnoja", "Nicolas Heess"], "categories": ["cs.RO", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2022-04-12", "url": "https://arxiv.org/abs/2204.05893", "pdf_url": "https://arxiv.org/pdf/2204.05893v2", "arxiv_id": "2204.05893", "doi": null, "citation_count": 8, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2386}
{"id": "04f329fbe4fb56eb1f4dffda2a3f2f667deea7864d8ea43c400673c5691f17c2", "sources": ["arxiv", "semantic_scholar"], "title": "A Closer Look at Rehearsal-Free Continual Learning", "abstract": "Continual learning is a setting where machine learning models learn novel concepts from continuously shifting training data, while simultaneously avoiding degradation of knowledge on previously seen classes which may disappear from the training data for extended periods of time (a phenomenon known as the catastrophic forgetting problem). Current approaches for continual learning of a single expanding task (aka class-incremental continual learning) require extensive rehearsal of previously seen data to avoid this degradation of knowledge. Unfortunately, rehearsal comes at a cost to memory, and it may also violate data-privacy. Instead, we explore combining knowledge distillation and parameter regularization in new ways to achieve strong continual learning performance without rehearsal. Specifically, we take a deep dive into common continual learning techniques: prediction distillation, feature distillation, L2 parameter regularization, and EWC parameter regularization. We first disprove the common assumption that parameter regularization techniques fail for rehearsal-free continual learning of a single, expanding task. Next, we explore how to leverage knowledge from a pre-trained model in rehearsal-free continual learning and find that vanilla L2 parameter regularization outperforms EWC parameter regularization and feature distillation. Finally, we explore the recently popular ImageNet-R benchmark, and show that L2 parameter regularization implemented in self-attention blocks of a ViT transformer outperforms recent popular prompting for continual learning methods.", "authors": ["James Seale Smith", "Junjiao Tian", "Shaunak Halbe", "Yen-Chang Hsu", "Zsolt Kira"], "categories": ["cs.LG", "cs.AI", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2022-03-31", "url": "https://arxiv.org/abs/2203.17269", "pdf_url": "https://arxiv.org/pdf/2203.17269v2", "arxiv_id": "2203.17269", "doi": "10.1109/CVPRW59228.2023.00239", "citation_count": 86, "influential_citation_count": 5, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.4849}
{"id": "258aa7d6092ff9741c5b7d5bc647ca8d94b64c91cc412cc6f2fead2b88440874", "sources": ["arxiv", "semantic_scholar"], "title": "Continual Normalization: Rethinking Batch Normalization for Online Continual Learning", "abstract": "Existing continual learning methods use Batch Normalization (BN) to facilitate training and improve generalization across tasks. However, the non-i.i.d and non-stationary nature of continual learning data, especially in the online setting, amplify the discrepancy between training and testing in BN and hinder the performance of older tasks. In this work, we study the cross-task normalization effect of BN in online continual learning where BN normalizes the testing data using moments biased towards the current task, resulting in higher catastrophic forgetting. This limitation motivates us to propose a simple yet effective method that we call Continual Normalization (CN) to facilitate training similar to BN while mitigating its negative effect. Extensive experiments on different continual learning algorithms and online scenarios show that CN is a direct replacement for BN and can provide substantial performance improvements. Our implementation is available at \\url{https://github.com/phquang/Continual-Normalization}.", "authors": ["Quang Pham", "Chenghao Liu", "Steven Hoi"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2022-03-30", "url": "https://arxiv.org/abs/2203.16102", "pdf_url": "https://arxiv.org/pdf/2203.16102v1", "arxiv_id": "2203.16102", "doi": "10.48550/arXiv.2203.16102", "citation_count": 74, "influential_citation_count": 8, "has_code": true, "code_url": "https://github.com/phquang/Continual-Normalization}", "venue": "International Conference on Learning Representations", "quality_score": 0.4771}
{"id": "eb00181bd45f657f9dbd423e5fd63394ced48078f9861bf3fb8d102d45a4d803", "sources": ["arxiv", "semantic_scholar"], "title": "Towards Exemplar-Free Continual Learning in Vision Transformers: an Account of Attention, Functional and Weight Regularization", "abstract": "In this paper, we investigate the continual learning of Vision Transformers (ViT) for the challenging exemplar-free scenario, with special focus on how to efficiently distill the knowledge of its crucial self-attention mechanism (SAM). Our work takes an initial step towards a surgical investigation of SAM for designing coherent continual learning methods in ViTs. We first carry out an evaluation of established continual learning regularization techniques. We then examine the effect of regularization when applied to two key enablers of SAM: (a) the contextualized embedding layers, for their ability to capture well-scaled representations with respect to the values, and (b) the prescaled attention maps, for carrying value-independent global contextual information. We depict the perks of each distilling strategy on two image recognition benchmarks (CIFAR100 and ImageNet-32) -- while (a) leads to a better overall accuracy, (b) helps enhance the rigidity by maintaining competitive performances. Furthermore, we identify the limitation imposed by the symmetric nature of regularization losses. To alleviate this, we propose an asymmetric variant and apply it to the pooled output distillation (POD) loss adapted for ViTs. Our experiments confirm that introducing asymmetry to POD boosts its plasticity while retaining stability across (a) and (b). Moreover, we acknowledge low forgetting measures for all the compared methods, indicating that ViTs might be naturally inclined continual learner", "authors": ["Francesco Pelosin", "Saurav Jha", "Andrea Torsello", "Bogdan Raducanu", "Joost van de Weijer"], "categories": ["cs.CV", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2022-03-24", "url": "https://arxiv.org/abs/2203.13167", "pdf_url": "https://arxiv.org/pdf/2203.13167v4", "arxiv_id": "2203.13167", "doi": "10.1109/CVPRW56347.2022.00427", "citation_count": 41, "influential_citation_count": 3, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.4058}
{"id": "926a6d14949d83ac5fc5053aac82a00d71a34903e5ee29a21d12a45f3048fc55", "sources": ["arxiv", "semantic_scholar"], "title": "Continual Learning and Private Unlearning", "abstract": "As intelligent agents become autonomous over longer periods of time, they may eventually become lifelong counterparts to specific people. If so, it may be common for a user to want the agent to master a task temporarily but later on to forget the task due to privacy concerns. However enabling an agent to \\emph{forget privately} what the user specified without degrading the rest of the learned knowledge is a challenging problem. With the aim of addressing this challenge, this paper formalizes this continual learning and private unlearning (CLPU) problem. The paper further introduces a straightforward but exactly private solution, CLPU-DER++, as the first step towards solving the CLPU problem, along with a set of carefully designed benchmark problems to evaluate the effectiveness of the proposed solution. The code is available at https://github.com/Cranial-XIX/Continual-Learning-Private-Unlearning.", "authors": ["Bo Liu", "Qiang Liu", "Peter Stone"], "categories": ["cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2022-03-24", "url": "https://arxiv.org/abs/2203.12817", "pdf_url": "https://arxiv.org/pdf/2203.12817v2", "arxiv_id": "2203.12817", "doi": "10.48550/arXiv.2203.12817", "citation_count": 147, "influential_citation_count": 20, "has_code": true, "code_url": "https://github.com/Cranial-XIX/Continual-Learning-Private-Unlearning", "venue": null, "quality_score": 0.6611}
{"id": "49d67a3b43d4b2a66f2434805ea55527f0c90c8296dd7f707c1a9b8f5776cdee", "sources": ["arxiv", "semantic_scholar"], "title": "Probing Representation Forgetting in Supervised and Unsupervised Continual Learning", "abstract": "Continual Learning research typically focuses on tackling the phenomenon of catastrophic forgetting in neural networks. Catastrophic forgetting is associated with an abrupt loss of knowledge previously learned by a model when the task, or more broadly the data distribution, being trained on changes. In supervised learning problems this forgetting, resulting from a change in the model's representation, is typically measured or observed by evaluating the decrease in old task performance. However, a model's representation can change without losing knowledge about prior tasks. In this work we consider the concept of representation forgetting, observed by using the difference in performance of an optimal linear classifier before and after a new task is introduced. Using this tool we revisit a number of standard continual learning benchmarks and observe that, through this lens, model representations trained without any explicit control for forgetting often experience small representation forgetting and can sometimes be comparable to methods which explicitly control for forgetting, especially in longer task sequences. We also show that representation forgetting can lead to new insights on the effect of model capacity and loss function used in continual learning. Based on our results, we show that a simple yet competitive approach is to learn representations continually with standard supervised contrastive learning while constructing prototypes of class samples when queried on old samples.", "authors": ["MohammadReza Davari", "Nader Asadi", "Sudhir Mudur", "Rahaf Aljundi", "Eugene Belilovsky"], "categories": ["cs.LG", "cs.AI", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2022-03-24", "url": "https://arxiv.org/abs/2203.13381", "pdf_url": "https://arxiv.org/pdf/2203.13381v2", "arxiv_id": "2203.13381", "doi": "10.1109/CVPR52688.2022.01621", "citation_count": 97, "influential_citation_count": 7, "has_code": false, "code_url": null, "venue": "Computer Vision and Pattern Recognition", "quality_score": 0.4978}
{"id": "f4f0fe525d733b45b1c49c6f45e1a93f25d46d1d1333c28573ef881c624a220a", "sources": ["arxiv", "semantic_scholar"], "title": "Online Continual Learning for Embedded Devices", "abstract": "Real-time on-device continual learning is needed for new applications such as home robots, user personalization on smartphones, and augmented/virtual reality headsets. However, this setting poses unique challenges: embedded devices have limited memory and compute capacity and conventional machine learning models suffer from catastrophic forgetting when updated on non-stationary data streams. While several online continual learning models have been developed, their effectiveness for embedded applications has not been rigorously studied. In this paper, we first identify criteria that online continual learners must meet to effectively perform real-time, on-device learning. We then study the efficacy of several online continual learning methods when used with mobile neural networks. We measure their performance, memory usage, compute requirements, and ability to generalize to out-of-domain inputs.", "authors": ["Tyler L. Hayes", "Christopher Kanan"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2022-03-21", "url": "https://arxiv.org/abs/2203.10681", "pdf_url": "https://arxiv.org/pdf/2203.10681v3", "arxiv_id": "2203.10681", "doi": "10.48550/arXiv.2203.10681", "citation_count": 72, "influential_citation_count": 9, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.5}
{"id": "98b7506ea76780551ce2aced32c5d9184938a2b4e309ad731114637ef784e730", "sources": ["arxiv", "semantic_scholar"], "title": "A Framework and Benchmark for Deep Batch Active Learning for Regression", "abstract": "The acquisition of labels for supervised learning can be expensive. To improve the sample efficiency of neural network regression, we study active learning methods that adaptively select batches of unlabeled data for labeling. We present a framework for constructing such methods out of (network-dependent) base kernels, kernel transformations, and selection methods. Our framework encompasses many existing Bayesian methods based on Gaussian process approximations of neural networks as well as non-Bayesian methods. Additionally, we propose to replace the commonly used last-layer features with sketched finite-width neural tangent kernels and to combine them with a novel clustering method. To evaluate different methods, we introduce an open-source benchmark consisting of 15 large tabular regression data sets. Our proposed method outperforms the state-of-the-art on our benchmark, scales to large data sets, and works out-of-the-box without adjusting the network architecture or training code. We provide open-source code that includes efficient implementations of all kernels, kernel transformations, and selection methods, and can be used for reproducing our results.", "authors": ["David Holzmüller", "Viktor Zaverkin", "Johannes Kästner", "Ingo Steinwart"], "categories": ["stat.ML", "cs.LG", "cs.NE"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2022-03-17", "url": "https://arxiv.org/abs/2203.09410", "pdf_url": "https://arxiv.org/pdf/2203.09410v4", "arxiv_id": "2203.09410", "doi": "10.48550/arXiv.2203.09410", "citation_count": 63, "influential_citation_count": 12, "has_code": true, "code_url": "https://github.com/dholzmueller/bmdal_reg", "venue": "Journal of machine learning research", "quality_score": 0.557}
{"id": "6fdcf04c181a83d1a71c26431f9a9bfb2f5a55f91db79d27a91203ac108e811f", "sources": ["arxiv", "semantic_scholar"], "title": "Overcoming Catastrophic Forgetting beyond Continual Learning: Balanced Training for Neural Machine Translation", "abstract": "Neural networks tend to gradually forget the previously learned knowledge when learning multiple tasks sequentially from dynamic data distributions. This problem is called \\textit{catastrophic forgetting}, which is a fundamental challenge in the continual learning of neural networks. In this work, we observe that catastrophic forgetting not only occurs in continual learning but also affects the traditional static training. Neural networks, especially neural machine translation models, suffer from catastrophic forgetting even if they learn from a static training set. To be specific, the final model pays imbalanced attention to training samples, where recently exposed samples attract more attention than earlier samples. The underlying cause is that training samples do not get balanced training in each model update, so we name this problem \\textit{imbalanced training}. To alleviate this problem, we propose Complementary Online Knowledge Distillation (COKD), which uses dynamically updated teacher models trained on specific data orders to iteratively provide complementary knowledge to the student model. Experimental results on multiple machine translation tasks show that our method successfully alleviates the problem of imbalanced training and achieves substantial improvements over strong baseline systems.", "authors": ["Chenze Shao", "Yang Feng"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2022-03-08", "url": "https://arxiv.org/abs/2203.03910", "pdf_url": "https://arxiv.org/pdf/2203.03910v2", "arxiv_id": "2203.03910", "doi": "10.48550/arXiv.2203.03910", "citation_count": 42, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": "Annual Meeting of the Association for Computational Linguistics", "quality_score": 0.4084}
{"id": "cdcde6cf2870ca46be96609c1ee4f407013cace4f77683b4c147a5bf595c8ce3", "sources": ["arxiv", "semantic_scholar"], "title": "Graph Neural Networks for Image Classification and Reinforcement Learning using Graph representations", "abstract": "In this paper, we will evaluate the performance of graph neural networks in two distinct domains: computer vision and reinforcement learning. In the computer vision section, we seek to learn whether a novel non-redundant representation for images as graphs can improve performance over trivial pixel to node mapping on a graph-level prediction graph, specifically image classification. For the reinforcement learning section, we seek to learn if explicitly modeling solving a Rubik's cube as a graph problem can improve performance over a standard model-free technique with no inductive bias.", "authors": ["Naman Goyal", "David Steiner"], "categories": ["cs.LG", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2022-03-07", "url": "https://arxiv.org/abs/2203.03457", "pdf_url": "https://arxiv.org/pdf/2203.03457v2", "arxiv_id": "2203.03457", "doi": "10.48550/arXiv.2203.03457", "citation_count": 5, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1945}
{"id": "ce04e1ee3c3336291dd3614484d869086ec7289497a91042c71dd5737cff6aa9", "sources": ["arxiv", "semantic_scholar"], "title": "Acceleration of Federated Learning with Alleviated Forgetting in Local Training", "abstract": "Federated learning (FL) enables distributed optimization of machine learning models while protecting privacy by independently training local models on each client and then aggregating parameters on a central server, thereby producing an effective global model. Although a variety of FL algorithms have been proposed, their training efficiency remains low when the data are not independently and identically distributed (non-i.i.d.) across different clients. We observe that the slow convergence rates of the existing methods are (at least partially) caused by the catastrophic forgetting issue during the local training stage on each individual client, which leads to a large increase in the loss function concerning the previous training data at the other clients. Here, we propose FedReg, an algorithm to accelerate FL with alleviated knowledge forgetting in the local training stage by regularizing locally trained parameters with the loss on generated pseudo data, which encode the knowledge of previous training data learned by the global model. Our comprehensive experiments demonstrate that FedReg not only significantly improves the convergence rate of FL, especially when the neural network architecture is deep and the clients' data are extremely non-i.i.d., but is also able to protect privacy better in classification problems and more robust against gradient inversion attacks. The code is available at: https://github.com/Zoesgithub/FedReg.", "authors": ["Chencheng Xu", "Zhiwei Hong", "Minlie Huang", "Tao Jiang"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2022-03-05", "url": "https://arxiv.org/abs/2203.02645", "pdf_url": "https://arxiv.org/pdf/2203.02645v1", "arxiv_id": "2203.02645", "doi": "10.48550/arXiv.2203.02645", "citation_count": 64, "influential_citation_count": 4, "has_code": true, "code_url": "https://github.com/Zoesgithub/FedReg", "venue": "International Conference on Learning Representations", "quality_score": 0.4532}
{"id": "b8c1f22aaa4fbaa87adbad941c7cfdb1d8c0da2ff340a760ff13e02023b72f5c", "sources": ["arxiv", "semantic_scholar"], "title": "Continual Learning Beyond a Single Model", "abstract": "A growing body of research in continual learning focuses on the catastrophic forgetting problem. While many attempts have been made to alleviate this problem, the majority of the methods assume a single model in the continual learning setup. In this work, we question this assumption and show that employing ensemble models can be a simple yet effective method to improve continual performance. However, ensembles' training and inference costs can increase significantly as the number of models grows. Motivated by this limitation, we study different ensemble models to understand their benefits and drawbacks in continual learning scenarios. Finally, to overcome the high compute cost of ensembles, we leverage recent advances in neural network subspace to propose a computationally cheap algorithm with similar runtime to a single model yet enjoying the performance benefits of ensembles.", "authors": ["Thang Doan", "Seyed Iman Mirzadeh", "Mehrdad Farajtabar"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2022-02-20", "url": "https://arxiv.org/abs/2202.09826", "pdf_url": "https://arxiv.org/pdf/2202.09826v3", "arxiv_id": "2202.09826", "doi": null, "citation_count": 24, "influential_citation_count": 3, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.3495}
{"id": "b8e133823ab53af787082b27f43cd8f2e47e8383973e5016ca4a0e334d6e5078", "sources": ["arxiv", "semantic_scholar"], "title": "A Neural Network Model of Continual Learning with Cognitive Control", "abstract": "Neural networks struggle in continual learning settings from catastrophic forgetting: when trials are blocked, new learning can overwrite the learning from previous blocks. Humans learn effectively in these settings, in some cases even showing an advantage of blocking, suggesting the brain contains mechanisms to overcome this problem. Here, we build on previous work and show that neural networks equipped with a mechanism for cognitive control do not exhibit catastrophic forgetting when trials are blocked. We further show an advantage of blocking over interleaving when there is a bias for active maintenance in the control signal, implying a tradeoff between maintenance and the strength of control. Analyses of map-like representations learned by the networks provided additional insights into these mechanisms. Our work highlights the potential of cognitive control to aid continual learning in neural networks, and offers an explanation for the advantage of blocking that has been observed in humans.", "authors": ["Jacob Russin", "Maryam Zolfaghar", "Seongmin A. Park", "Erie Boorman", "Randall C. O'Reilly"], "categories": ["q-bio.NC", "cs.LG", "cs.NE"], "fields_of_study": ["Computer Science", "Biology", "Medicine"], "published_date": "2022-02-09", "url": "https://arxiv.org/abs/2202.04773", "pdf_url": "https://arxiv.org/pdf/2202.04773v2", "arxiv_id": "2202.04773", "doi": null, "citation_count": 15, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": "Annual Meeting of the Cognitive Science Society", "quality_score": 0.301}
{"id": "d38959ed162e3866161eae0f48ccd74ebf1b12dd916faa3e9d4f8eafd358a17a", "sources": ["arxiv", "semantic_scholar"], "title": "Learning Curves for Decision Making in Supervised Machine Learning: A Survey", "abstract": "Learning curves are a concept from social sciences that has been adopted in the context of machine learning to assess the performance of a learning algorithm with respect to a certain resource, e.g., the number of training examples or the number of training iterations. Learning curves have important applications in several machine learning contexts, most notably in data acquisition, early stopping of model training, and model selection. For instance, learning curves can be used to model the performance of the combination of an algorithm and its hyperparameter configuration, providing insights into their potential suitability at an early stage and often expediting the algorithm selection process. Various learning curve models have been proposed to use learning curves for decision making. Some of these models answer the binary decision question of whether a given algorithm at a certain budget will outperform a certain reference performance, whereas more complex models predict the entire learning curve of an algorithm. We contribute a framework that categorises learning curve approaches using three criteria: the decision-making situation they address, the intrinsic learning curve question they answer and the type of resources they use. We survey papers from the literature and classify them into this framework.", "authors": ["Felix Mohr", "Jan N. van Rijn"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2022-01-28", "url": "https://arxiv.org/abs/2201.12150", "pdf_url": "https://arxiv.org/pdf/2201.12150v2", "arxiv_id": "2201.12150", "doi": "10.1007/s10994-024-06619-7", "citation_count": 90, "influential_citation_count": 5, "has_code": false, "code_url": null, "venue": "Machine-mediated learning", "quality_score": 0.4898}
{"id": "83f4e60da703bac4a360206f1d0184f046f1fc4aaa272d5e9c410914caba78b7", "sources": ["arxiv", "semantic_scholar"], "title": "Combining Optimal Path Search With Task-Dependent Learning in a Neural Network", "abstract": "Finding optimal paths in connected graphs requires determining the smallest total cost for traveling along the graph's edges. This problem can be solved by several classical algorithms where, usually, costs are predefined for all edges. Conventional planning methods can, thus, normally not be used when wanting to change costs in an adaptive way following the requirements of some task. Here we show that one can define a neural network representation of path finding problems by transforming cost values into synaptic weights, which allows for online weight adaptation using network learning mechanisms. When starting with an initial activity value of one, activity propagation in this network will lead to solutions, which are identical to those found by the Bellman-Ford algorithm. The neural network has the same algorithmic complexity as Bellman-Ford and, in addition, we can show that network learning mechanisms (such as Hebbian learning) can adapt the weights in the network augmenting the resulting paths according to some task at hand. We demonstrate this by learning to navigate in an environment with obstacles as well as by learning to follow certain sequences of path nodes. Hence, the here-presented novel algorithm may open up a different regime of applications where path-augmentation (by learning) is directly coupled with path finding in a natural way.", "authors": ["Tomas Kulvicius", "Minija Tamosiunaite", "Florentin Wörgötter"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science", "Medicine"], "published_date": "2022-01-26", "url": "https://arxiv.org/abs/2201.11104", "pdf_url": "https://arxiv.org/pdf/2201.11104v6", "arxiv_id": "2201.11104", "doi": "10.1109/TNNLS.2023.3327103", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.0753}
{"id": "43e4af76da24024b8ba53ad43e5b3c68bb3fe6a734d4e12fe7aa2d1ae9aff7fa", "sources": ["arxiv", "semantic_scholar"], "title": "Visualizing the Diversity of Representations Learned by Bayesian Neural Networks", "abstract": "Explainable Artificial Intelligence (XAI) aims to make learning machines less opaque, and offers researchers and practitioners various tools to reveal the decision-making strategies of neural networks. In this work, we investigate how XAI methods can be used for exploring and visualizing the diversity of feature representations learned by Bayesian Neural Networks (BNNs). Our goal is to provide a global understanding of BNNs by making their decision-making strategies a) visible and tangible through feature visualizations and b) quantitatively measurable with a distance measure learned by contrastive learning. Our work provides new insights into the \\emph{posterior} distribution in terms of human-understandable feature information with regard to the underlying decision making strategies. The main findings of our work are the following: 1) global XAI methods can be applied to explain the diversity of decision-making strategies of BNN instances, 2) Monte Carlo dropout with commonly used Dropout rates exhibit increased diversity in feature representations compared to the multimodal posterior approximation of MultiSWAG, 3) the diversity of learned feature representations highly correlates with the uncertainty estimate for the output and 4) the inter-mode diversity of the multimodal posterior decreases as the network width increases, while the intra mode diversity increases. These findings are consistent with the recent Deep Neural Networks theory, providing additional intuitions about what the theory implies in terms of humanly understandable concepts.", "authors": ["Dennis Grinwald", "Kirill Bykov", "Shinichi Nakajima", "Marina M. -C. Höhne"], "categories": ["cs.LG", "cs.AI", "cs.CV", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2022-01-26", "url": "https://arxiv.org/abs/2201.10859", "pdf_url": "https://arxiv.org/pdf/2201.10859v2", "arxiv_id": "2201.10859", "doi": null, "citation_count": 7, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Published in Transactions on Machine Learning Research (11/2023)", "quality_score": 0.2258}
{"id": "f5371aef333465fc855a1243b6b34b6ae193fa9a1348c1ab15636b0c95bb5717", "sources": ["arxiv", "semantic_scholar"], "title": "Learning to Predict Gradients for Semi-Supervised Continual Learning", "abstract": "A key challenge for machine intelligence is to learn new visual concepts without forgetting the previously acquired knowledge. Continual learning is aimed towards addressing this challenge. However, there is a gap between existing supervised continual learning and human-like intelligence, where human is able to learn from both labeled and unlabeled data. How unlabeled data affects learning and catastrophic forgetting in the continual learning process remains unknown. To explore these issues, we formulate a new semi-supervised continual learning method, which can be generically applied to existing continual learning models. Specifically, a novel gradient learner learns from labeled data to predict gradients on unlabeled data. Hence, the unlabeled data could fit into the supervised continual learning method. Different from conventional semi-supervised settings, we do not hypothesize that the underlying classes, which are associated to the unlabeled data, are known to the learning process. In other words, the unlabeled data could be very distinct from the labeled data. We evaluate the proposed method on mainstream continual learning, adversarial continual learning, and semi-supervised learning tasks. The proposed method achieves state-of-the-art performance on classification accuracy and backward transfer in the continual learning setting while achieving desired performance on classification accuracy in the semi-supervised learning setting. This implies that the unlabeled images can enhance the generalizability of continual learning models on the predictive ability on unseen data and significantly alleviate catastrophic forgetting. The code is available at \\url{https://github.com/luoyan407/grad_prediction.git}.", "authors": ["Yan Luo", "Yongkang Wong", "Mohan Kankanhalli", "Qi Zhao"], "categories": ["cs.LG", "cs.CV"], "fields_of_study": ["Computer Science", "Medicine"], "published_date": "2022-01-23", "url": "https://arxiv.org/abs/2201.09196", "pdf_url": "https://arxiv.org/pdf/2201.09196v2", "arxiv_id": "2201.09196", "doi": "10.1109/TNNLS.2024.3361375", "citation_count": 14, "influential_citation_count": 2, "has_code": true, "code_url": "https://github.com/luoyan407/grad_prediction.git}", "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.294}
{"id": "9309167c5db651ec09bbea94c7518cb59df6ce1813e2bc72ae4cfacc1239328e", "sources": ["arxiv", "semantic_scholar"], "title": "Automatic Sparse Connectivity Learning for Neural Networks", "abstract": "Since sparse neural networks usually contain many zero weights, these unnecessary network connections can potentially be eliminated without degrading network performance. Therefore, well-designed sparse neural networks have the potential to significantly reduce FLOPs and computational resources. In this work, we propose a new automatic pruning method - Sparse Connectivity Learning (SCL). Specifically, a weight is re-parameterized as an element-wise multiplication of a trainable weight variable and a binary mask. Thus, network connectivity is fully described by the binary mask, which is modulated by a unit step function. We theoretically prove the fundamental principle of using a straight-through estimator (STE) for network pruning. This principle is that the proxy gradients of STE should be positive, ensuring that mask variables converge at their minima. After finding Leaky ReLU, Softplus, and Identity STEs can satisfy this principle, we propose to adopt Identity STE in SCL for discrete mask relaxation. We find that mask gradients of different features are very unbalanced, hence, we propose to normalize mask gradients of each feature to optimize mask variable training. In order to automatically train sparse masks, we include the total number of network connections as a regularization term in our objective function. As SCL does not require pruning criteria or hyper-parameters defined by designers for network layers, the network is explored in a larger hypothesis space to achieve optimized sparse connectivity for the best performance. SCL overcomes the limitations of existing automatic pruning methods. Experimental results demonstrate that SCL can automatically learn and select important network connections for various baseline network structures. Deep learning models trained by SCL outperform the SOTA human-designed and automatic pruning methods in sparsity, accuracy, and FLOPs reduction.", "authors": ["Zhimin Tang", "Linkai Luo", "Bike Xie", "Yiyu Zhu", "Rujie Zhao", "Lvqing Bi", "Chao Lu"], "categories": ["cs.CV", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science", "Medicine"], "published_date": "2022-01-13", "url": "https://arxiv.org/abs/2201.05020", "pdf_url": "https://arxiv.org/pdf/2201.05020v1", "arxiv_id": "2201.05020", "doi": "10.1109/TNNLS.2022.3141665", "citation_count": 50, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.4269}
{"id": "b75dab3e845cb4527ca0679885582cc2e75c5e595416b4893e3eec4116b0a7b5", "sources": ["arxiv", "semantic_scholar"], "title": "Continually Learning Self-Supervised Representations with Projected Functional Regularization", "abstract": "Recent self-supervised learning methods are able to learn high-quality image representations and are closing the gap with supervised approaches. However, these methods are unable to acquire new knowledge incrementally -- they are, in fact, mostly used only as a pre-training phase over IID data. In this work we investigate self-supervised methods in continual learning regimes without any replay mechanism. We show that naive functional regularization, also known as feature distillation, leads to lower plasticity and limits continual learning performance. Instead, we propose Projected Functional Regularization in which a separate temporal projection network ensures that the newly learned feature space preserves information of the previous one, while at the same time allowing for the learning of new features. This prevents forgetting while maintaining the plasticity of the learner. Comparison with other incremental learning approaches applied to self-supervision demonstrates that our method obtains competitive performance in different scenarios and on multiple datasets.", "authors": ["Alex Gomez-Villa", "Bartlomiej Twardowski", "Lu Yu", "Andrew D. Bagdanov", "Joost van de Weijer"], "categories": ["cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2021-12-30", "url": "https://arxiv.org/abs/2112.15022", "pdf_url": "https://arxiv.org/pdf/2112.15022v2", "arxiv_id": "2112.15022", "doi": "10.1109/CVPRW56347.2022.00432", "citation_count": 55, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.437}
{"id": "5949321441070c37c5f3c6bec6b5f2b2b0e2782d23aaa7c3d637601c7764e85d", "sources": ["arxiv"], "title": "Federated Learning with Superquantile Aggregation for Heterogeneous Data", "abstract": "We present a federated learning framework that is designed to robustly deliver good predictive performance across individual clients with heterogeneous data. The proposed approach hinges upon a superquantile-based learning objective that captures the tail statistics of the error distribution over heterogeneous clients. We present a stochastic training algorithm that interleaves differentially private client filtering with federated averaging steps. We prove finite time convergence guarantees for the algorithm: $O(1/\\sqrt{T})$ in the nonconvex case in $T$ communication rounds and $O(\\exp(-T/κ^{3/2}) + κ/T)$ in the strongly convex case with local condition number $κ$. Experimental results on benchmark datasets for federated learning demonstrate that our approach is competitive with classical ones in terms of average error and outperforms them in terms of tail statistics of the error.", "authors": ["Krishna Pillutla", "Yassine Laguel", "Jérôme Malick", "Zaid Harchaoui"], "categories": ["cs.LG", "math.OC", "stat.ML"], "fields_of_study": [], "published_date": "2021-12-17", "url": "https://arxiv.org/abs/2112.09429", "pdf_url": "https://arxiv.org/pdf/2112.09429v2", "arxiv_id": "2112.09429", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Machine Learning (2023): 1-68", "quality_score": 0.0}
{"id": "8c300fed7da60ed87352081870afad00617caa2d2ba3d1adafd6d94538411557", "sources": ["arxiv", "semantic_scholar"], "title": "An Empirical Investigation of the Role of Pre-training in Lifelong Learning", "abstract": "The lifelong learning paradigm in machine learning is an attractive alternative to the more prominent isolated learning scheme not only due to its resemblance to biological learning but also its potential to reduce energy waste by obviating excessive model re-training. A key challenge to this paradigm is the phenomenon of catastrophic forgetting. With the increasing popularity and success of pre-trained models in machine learning, we pose the question: What role does pre-training play in lifelong learning, specifically with respect to catastrophic forgetting? We investigate existing methods in the context of large, pre-trained models and evaluate their performance on a variety of text and image classification tasks, including a large-scale study using a novel data set of 15 diverse NLP tasks. Across all settings, we observe that generic pre-training implicitly alleviates the effects of catastrophic forgetting when learning multiple tasks sequentially compared to randomly initialized models. We then further investigate why pre-training alleviates forgetting in this setting. We study this phenomenon by analyzing the loss landscape, finding that pre-trained weights appear to ease forgetting by leading to wider minima. Based on this insight, we propose jointly optimizing for current task loss and loss basin sharpness to explicitly encourage wider basins during sequential fine-tuning. We show that this optimization approach outperforms several state-of-the-art task-sequential continual learning algorithms across multiple settings, occasionally even without retaining a memory that scales in size with the number of tasks.", "authors": ["Sanket Vaibhav Mehta", "Darshan Patil", "Sarath Chandar", "Emma Strubell"], "categories": ["cs.LG", "cs.AI", "cs.CL", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2021-12-16", "url": "https://arxiv.org/abs/2112.09153", "pdf_url": "https://arxiv.org/pdf/2112.09153v2", "arxiv_id": "2112.09153", "doi": null, "citation_count": 178, "influential_citation_count": 5, "has_code": false, "code_url": null, "venue": "Journal of machine learning research", "quality_score": 0.5632}
{"id": "8d6f0008cee3a14d8e085fefd83b387d1b845cd84a80dd255bb7767d98c14fbc", "sources": ["arxiv", "semantic_scholar"], "title": "CSG0: Continual Urban Scene Generation with Zero Forgetting", "abstract": "With the rapid advances in generative adversarial networks (GANs), the visual quality of synthesised scenes keeps improving, including for complex urban scenes with applications to automated driving. We address in this work a continual scene generation setup in which GANs are trained on a stream of distinct domains; ideally, the learned models should eventually be able to generate new scenes in all seen domains. This setup reflects the real-life scenario where data are continuously acquired in different places at different times. In such a continual setup, we aim for learning with zero forgetting, \\IE, with no degradation in synthesis quality over earlier domains due to catastrophic forgetting. To this end, we introduce a novel framework that not only (i) enables seamless knowledge transfer in continual training but also (ii) guarantees zero forgetting with a small overhead cost. While being more memory efficient, thanks to continual learning, our model obtains better synthesis quality as compared against the brute-force solution that trains one full model for each domain. Especially, under extreme low-data regimes, our approach outperforms the brute-force one by a large margin.", "authors": ["Himalaya Jain", "Tuan-Hung Vu", "Patrick Pérez", "Matthieu Cord"], "categories": ["cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2021-12-06", "url": "https://arxiv.org/abs/2112.03252", "pdf_url": "https://arxiv.org/pdf/2112.03252v2", "arxiv_id": "2112.03252", "doi": "10.1109/CVPRW56347.2022.00412", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.0}
{"id": "d329e034c642241f31160980318568097c73584846fc73c871875987b98e2af8", "sources": ["arxiv", "semantic_scholar"], "title": "Learning Curves for Continual Learning in Neural Networks: Self-Knowledge Transfer and Forgetting", "abstract": "Sequential training from task to task is becoming one of the major objects in deep learning applications such as continual learning and transfer learning. Nevertheless, it remains unclear under what conditions the trained model's performance improves or deteriorates. To deepen our understanding of sequential training, this study provides a theoretical analysis of generalization performance in a solvable case of continual learning. We consider neural networks in the neural tangent kernel (NTK) regime that continually learn target functions from task to task, and investigate the generalization by using an established statistical mechanical analysis of kernel ridge-less regression. We first show characteristic transitions from positive to negative transfer. More similar targets above a specific critical value can achieve positive knowledge transfer for the subsequent task while catastrophic forgetting occurs even with very similar targets. Next, we investigate a variant of continual learning which supposes the same target function in multiple tasks. Even for the same target, the trained model shows some transfer and forgetting depending on the sample size of each task. We can guarantee that the generalization error monotonically decreases from task to task for equal sample sizes while unbalanced sample sizes deteriorate the generalization. We respectively refer to these improvement and deterioration as self-knowledge transfer and forgetting, and empirically confirm them in realistic training of deep neural networks as well.", "authors": ["Ryo Karakida", "Shotaro Akaho"], "categories": ["stat.ML", "cond-mat.dis-nn", "cs.LG"], "fields_of_study": ["Computer Science", "Mathematics", "Physics"], "published_date": "2021-12-03", "url": "https://arxiv.org/abs/2112.01653", "pdf_url": "https://arxiv.org/pdf/2112.01653v2", "arxiv_id": "2112.01653", "doi": null, "citation_count": 16, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.3076}
{"id": "b8110a0c62fda8dc7b5c0d9f143a400561636ce491564a46ae0423d52a0f722a", "sources": ["arxiv", "semantic_scholar"], "title": "Learning by Active Forgetting for Neural Networks", "abstract": "Remembering and forgetting mechanisms are two sides of the same coin in a human learning-memory system. Inspired by human brain memory mechanisms, modern machine learning systems have been working to endow machine with lifelong learning capability through better remembering while pushing the forgetting as the antagonist to overcome. Nevertheless, this idea might only see the half picture. Up until very recently, increasing researchers argue that a brain is born to forget, i.e., forgetting is a natural and active process for abstract, rich, and flexible representations. This paper presents a learning model by active forgetting mechanism with artificial neural networks. The active forgetting mechanism (AFM) is introduced to a neural network via a \"plug-and-play\" forgetting layer (P\\&PF), consisting of groups of inhibitory neurons with Internal Regulation Strategy (IRS) to adjust the extinction rate of themselves via lateral inhibition mechanism and External Regulation Strategy (ERS) to adjust the extinction rate of excitatory neurons via inhibition mechanism. Experimental studies have shown that the P\\&PF offers surprising benefits: self-adaptive structure, strong generalization, long-term learning and memory, and robustness to data and parameter perturbation. This work sheds light on the importance of forgetting in the learning process and offers new perspectives to understand the underlying mechanisms of neural networks.", "authors": ["Jian Peng", "Xian Sun", "Min Deng", "Chao Tao", "Bo Tang", "Wenbo Li", "Guohua Wu", " QingZhu", "Yu Liu", "Tao Lin", "Haifeng Li"], "categories": ["cs.LG", "cs.AI", "cs.NE"], "fields_of_study": ["Computer Science"], "published_date": "2021-11-21", "url": "https://arxiv.org/abs/2111.10831", "pdf_url": "https://arxiv.org/pdf/2111.10831v1", "arxiv_id": "2111.10831", "doi": null, "citation_count": 4, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1747}
{"id": "78a197446f10b90defbdf191e328e2721f86fb030985033d480550a4ff2bf8b0", "sources": ["arxiv", "semantic_scholar"], "title": "Lifelong Learning from Event-based Data", "abstract": "Lifelong learning is a long-standing aim for artificial agents that act in dynamic environments, in which an agent needs to accumulate knowledge incrementally without forgetting previously learned representations. We investigate methods for learning from data produced by event cameras and compare techniques to mitigate forgetting while learning incrementally. We propose a model that is composed of both, feature extraction and continuous learning. Furthermore, we introduce a habituation-based method to mitigate forgetting. Our experimental results show that the combination of different techniques can help to avoid catastrophic forgetting while learning incrementally from the features provided by the extraction module.", "authors": ["Vadym Gryshchuk", "Cornelius Weber", "Chu Kiong Loo", "Stefan Wermter"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2021-11-11", "url": "https://arxiv.org/abs/2111.08458", "pdf_url": "https://arxiv.org/pdf/2111.08458v1", "arxiv_id": "2111.08458", "doi": "10.14428/esann/2021.es2021-146", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "The European Symposium on Artificial Neural Networks", "quality_score": 0.0}
{"id": "58d2ea335e1e4ab6ce4f983f802ea97b84f99a8cc6c94ec10404bcfa03714eca", "sources": ["arxiv", "semantic_scholar"], "title": "d3rlpy: An Offline Deep Reinforcement Learning Library", "abstract": "In this paper, we introduce d3rlpy, an open-sourced offline deep reinforcement learning (RL) library for Python. d3rlpy supports a set of offline deep RL algorithms as well as off-policy online algorithms via a fully documented plug-and-play API. To address a reproducibility issue, we conduct a large-scale benchmark with D4RL and Atari 2600 dataset to ensure implementation quality and provide experimental scripts and full tables of results. The d3rlpy source code can be found on GitHub: \\url{https://github.com/takuseno/d3rlpy}.", "authors": ["Takuma Seno", "Michita Imai"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2021-11-06", "url": "https://arxiv.org/abs/2111.03788", "pdf_url": "https://arxiv.org/pdf/2111.03788v2", "arxiv_id": "2111.03788", "doi": null, "citation_count": 144, "influential_citation_count": 10, "has_code": true, "code_url": "https://github.com/takuseno/d3rlpy}", "venue": "Journal of machine learning research", "quality_score": 0.5403}
{"id": "44ca0c5cd2838a2f6b6a511bd8d229e0aa97abff573077d0d06b70b6903771dc", "sources": ["arxiv", "semantic_scholar"], "title": "AFEC: Active Forgetting of Negative Transfer in Continual Learning", "abstract": "Continual learning aims to learn a sequence of tasks from dynamic data distributions. Without accessing to the old training samples, knowledge transfer from the old tasks to each new task is difficult to determine, which might be either positive or negative. If the old knowledge interferes with the learning of a new task, i.e., the forward knowledge transfer is negative, then precisely remembering the old tasks will further aggravate the interference, thus decreasing the performance of continual learning. By contrast, biological neural networks can actively forget the old knowledge that conflicts with the learning of a new experience, through regulating the learning-triggered synaptic expansion and synaptic convergence. Inspired by the biological active forgetting, we propose to actively forget the old knowledge that limits the learning of new tasks to benefit continual learning. Under the framework of Bayesian continual learning, we develop a novel approach named Active Forgetting with synaptic Expansion-Convergence (AFEC). Our method dynamically expands parameters to learn each new task and then selectively combines them, which is formally consistent with the underlying mechanism of biological active forgetting. We extensively evaluate AFEC on a variety of continual learning benchmarks, including CIFAR-10 regression tasks, visual classification tasks and Atari reinforcement tasks, where AFEC effectively improves the learning of new tasks and achieves the state-of-the-art performance in a plug-and-play way.", "authors": ["Liyuan Wang", "Mingtian Zhang", "Zhongfan Jia", "Qian Li", "Chenglong Bao", "Kaisheng Ma", "Jun Zhu", "Yi Zhong"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2021-10-23", "url": "https://arxiv.org/abs/2110.12187", "pdf_url": "https://arxiv.org/pdf/2110.12187v2", "arxiv_id": "2110.12187", "doi": null, "citation_count": 120, "influential_citation_count": 13, "has_code": false, "code_url": null, "venue": "Neural Information Processing Systems", "quality_score": 0.5731}
{"id": "c0462459f0c8bbde63263fad20d2c8638061189c3b83cd8aa2b13e0a78e6cd99", "sources": ["arxiv", "semantic_scholar"], "title": "Wide Neural Networks Forget Less Catastrophically", "abstract": "A primary focus area in continual learning research is alleviating the \"catastrophic forgetting\" problem in neural networks by designing new algorithms that are more robust to the distribution shifts. While the recent progress in continual learning literature is encouraging, our understanding of what properties of neural networks contribute to catastrophic forgetting is still limited. To address this, instead of focusing on continual learning algorithms, in this work, we focus on the model itself and study the impact of \"width\" of the neural network architecture on catastrophic forgetting, and show that width has a surprisingly significant effect on forgetting. To explain this effect, we study the learning dynamics of the network from various perspectives such as gradient orthogonality, sparsity, and lazy training regime. We provide potential explanations that are consistent with the empirical results across different architectures and continual learning benchmarks.", "authors": ["Seyed Iman Mirzadeh", "Arslan Chaudhry", "Dong Yin", "Huiyi Hu", "Razvan Pascanu", "Dilan Gorur", "Mehrdad Farajtabar"], "categories": ["cs.LG", "cs.AI", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2021-10-21", "url": "https://arxiv.org/abs/2110.11526", "pdf_url": "https://arxiv.org/pdf/2110.11526v3", "arxiv_id": "2110.11526", "doi": null, "citation_count": 87, "influential_citation_count": 11, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.5396}
{"id": "d3644669f55c61283267f091c6e2d05097e175848e3851a18fc2f90ca2cd75d2", "sources": ["arxiv", "semantic_scholar"], "title": "Carousel Memory: Rethinking the Design of Episodic Memory for Continual Learning", "abstract": "Continual Learning (CL) is an emerging machine learning paradigm that aims to learn from a continuous stream of tasks without forgetting knowledge learned from the previous tasks. To avoid performance decrease caused by forgetting, prior studies exploit episodic memory (EM), which stores a subset of the past observed samples while learning from new non-i.i.d. data. Despite the promising results, since CL is often assumed to execute on mobile or IoT devices, the EM size is bounded by the small hardware memory capacity and makes it infeasible to meet the accuracy requirements for real-world applications. Specifically, all prior CL methods discard samples overflowed from the EM and can never retrieve them back for subsequent training steps, incurring loss of information that would exacerbate catastrophic forgetting. We explore a novel hierarchical EM management strategy to address the forgetting issue. In particular, in mobile and IoT devices, real-time data can be stored not just in high-speed RAMs but in internal storage devices as well, which offer significantly larger capacity than the RAMs. Based on this insight, we propose to exploit the abundant storage to preserve past experiences and alleviate the forgetting by allowing CL to efficiently migrate samples between memory and storage without being interfered by the slow access speed of the storage. We call it Carousel Memory (CarM). As CarM is complementary to existing CL methods, we conduct extensive evaluations of our method with seven popular CL methods and show that CarM significantly improves the accuracy of the methods across different settings by large margins in final average accuracy (up to 28.4%) while retaining the same training efficiency.", "authors": ["Soobee Lee", "Minindu Weerakoon", "Jonghyun Choi", "Minjia Zhang", "Di Wang", "Myeongjae Jeon"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2021-10-14", "url": "https://arxiv.org/abs/2110.07276", "pdf_url": "https://arxiv.org/pdf/2110.07276v3", "arxiv_id": "2110.07276", "doi": null, "citation_count": 2, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1193}
{"id": "657a6f766a7557c748d26e79c0441e847f9ef4e04ea5ad46bd60e956bd1c9ae8", "sources": ["arxiv", "semantic_scholar"], "title": "LFPT5: A Unified Framework for Lifelong Few-shot Language Learning Based on Prompt Tuning of T5", "abstract": "Existing approaches to lifelong language learning rely on plenty of labeled data for learning a new task, which is hard to obtain in most real scenarios. Considering that humans can continually learn new tasks from a handful of examples, we expect the models also to be able to generalize well on new few-shot tasks without forgetting the previous ones. In this work, we define this more challenging yet practical problem as Lifelong Few-shot Language Learning (LFLL) and propose a unified framework for it based on prompt tuning of T5. Our framework called LFPT5 takes full advantage of PT's strong few-shot learning ability, and simultaneously trains the model as a task solver and a data generator. Before learning a new domain of the same task type, LFPT5 generates pseudo (labeled) samples of previously learned domains, and later gets trained on those samples to alleviate forgetting of previous knowledge as it learns the new domain. In addition, a KL divergence loss is minimized to achieve label consistency between the previous and the current model. While adapting to a new task type, LFPT5 includes and tunes additional prompt embeddings for the new task. With extensive experiments, we demonstrate that LFPT5 can be applied to various different types of tasks and significantly outperform previous methods in different LFLL settings.", "authors": ["Chengwei Qin", "Shafiq Joty"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2021-10-14", "url": "https://arxiv.org/abs/2110.07298", "pdf_url": "https://arxiv.org/pdf/2110.07298v3", "arxiv_id": "2110.07298", "doi": null, "citation_count": 133, "influential_citation_count": 13, "has_code": true, "code_url": "https://github.com/qcwthu/Lifelong-Fewshot-Language-Learning", "venue": "International Conference on Learning Representations", "quality_score": 0.5731}
{"id": "445920c0865aa39b3b4559bd89fc1216d03a848db6132e1f9adb4c7a4275c6b9", "sources": ["arxiv", "semantic_scholar"], "title": "Avoiding Forgetting and Allowing Forward Transfer in Continual Learning via Sparse Networks", "abstract": "Using task-specific components within a neural network in continual learning (CL) is a compelling strategy to address the stability-plasticity dilemma in fixed-capacity models without access to past data. Current methods focus only on selecting a sub-network for a new task that reduces forgetting of past tasks. However, this selection could limit the forward transfer of relevant past knowledge that helps in future learning. Our study reveals that satisfying both objectives jointly is more challenging when a unified classifier is used for all classes of seen tasks-class-Incremental Learning (class-IL)-as it is prone to ambiguities between classes across tasks. Moreover, the challenge increases when the semantic similarity of classes across tasks increases. To address this challenge, we propose a new CL method, named AFAF, that aims to Avoid Forgetting and Allow Forward transfer in class-IL using fix-capacity models. AFAF allocates a sub-network that enables selective transfer of relevant knowledge to a new task while preserving past knowledge, reusing some of the previously allocated components to utilize the fixed-capacity, and addressing class-ambiguities when similarities exist. The experiments show the effectiveness of AFAF in providing models with multiple CL desirable properties, while outperforming state-of-the-art methods on various challenging benchmarks with different semantic similarities.", "authors": ["Ghada Sokar", "Decebal Constantin Mocanu", "Mykola Pechenizkiy"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2021-10-11", "url": "https://arxiv.org/abs/2110.05329", "pdf_url": "https://arxiv.org/pdf/2110.05329v3", "arxiv_id": "2110.05329", "doi": "10.1007/978-3-031-26409-2_6", "citation_count": 10, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2603}
{"id": "f5dc40b4b1f68bc61bafe9b3ebc6dca8c0c7e099ce273f2ac4eb68f4707696a8", "sources": ["arxiv", "semantic_scholar"], "title": "QTN-VQC: An End-to-End Learning framework for Quantum Neural Networks", "abstract": "The advent of noisy intermediate-scale quantum (NISQ) computers raises a crucial challenge to design quantum neural networks for fully quantum learning tasks. To bridge the gap, this work proposes an end-to-end learning framework named QTN-VQC, by introducing a trainable quantum tensor network (QTN) for quantum embedding on a variational quantum circuit (VQC). The architecture of QTN is composed of a parametric tensor-train network for feature extraction and a tensor product encoding for quantum embedding. We highlight the QTN for quantum embedding in terms of two perspectives: (1) we theoretically characterize QTN by analyzing its representation power of input features; (2) QTN enables an end-to-end parametric model pipeline, namely QTN-VQC, from the generation of quantum embedding to the output measurement. Our experiments on the MNIST dataset demonstrate the advantages of QTN for quantum embedding over other quantum embedding approaches.", "authors": ["Jun Qi", "Chao-Han Huck Yang", "Pin-Yu Chen"], "categories": ["quant-ph", "cs.AI", "cs.CL", "cs.CV", "cs.LG", "cs.NE"], "fields_of_study": ["Physics", "Computer Science"], "published_date": "2021-10-06", "url": "https://arxiv.org/abs/2110.03861", "pdf_url": "https://arxiv.org/pdf/2110.03861v3", "arxiv_id": "2110.03861", "doi": "10.1088/1402-4896/ad14d6", "citation_count": 65, "influential_citation_count": 4, "has_code": false, "code_url": null, "venue": "Physica Scripta", "quality_score": 0.4549}
{"id": "911493f35fdab11c6dfaa6bf450439429ea6b25aa44842e029e1a059a2b6d70f", "sources": ["arxiv", "semantic_scholar"], "title": "Assisted Learning for Organizations with Limited Imbalanced Data", "abstract": "In the era of big data, many big organizations are integrating machine learning into their work pipelines to facilitate data analysis. However, the performance of their trained models is often restricted by limited and imbalanced data available to them. In this work, we develop an assisted learning framework for assisting organizations to improve their learning performance. The organizations have sufficient computation resources but are subject to stringent data-sharing and collaboration policies. Their limited imbalanced data often cause biased inference and sub-optimal decision-making. In assisted learning, an organizational learner purchases assistance service from an external service provider and aims to enhance its model performance within only a few assistance rounds. We develop effective stochastic training algorithms for both assisted deep learning and assisted reinforcement learning. Different from existing distributed algorithms that need to frequently transmit gradients or models, our framework allows the learner to only occasionally share information with the service provider, but still obtain a model that achieves near-oracle performance as if all the data were centralized.", "authors": ["Cheng Chen", "Jiaying Zhou", "Jie Ding", "Yi Zhou"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2021-09-20", "url": "https://arxiv.org/abs/2109.09307", "pdf_url": "https://arxiv.org/pdf/2109.09307v4", "arxiv_id": "2109.09307", "doi": null, "citation_count": 3, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "C. Chen, J. Zhou, J. Ding, and Y. Zhou, \"Assisted Learning for Organizations with Limited Imbalanced Data,\" Transactions on Machine Learning Research (TMLR), 2023", "quality_score": 0.1505}
{"id": "1c12afc65a9eb52c71c6485436ca9f10430a7e020532bab66e9bd15f9602f466", "sources": ["arxiv", "semantic_scholar"], "title": "Concave Utility Reinforcement Learning with Zero-Constraint Violations", "abstract": "We consider the problem of tabular infinite horizon concave utility reinforcement learning (CURL) with convex constraints. For this, we propose a model-based learning algorithm that also achieves zero constraint violations. Assuming that the concave objective and the convex constraints have a solution interior to the set of feasible occupation measures, we solve a tighter optimization problem to ensure that the constraints are never violated despite the imprecise model knowledge and model stochasticity. We use Bellman error-based analysis for tabular infinite-horizon setups which allows analyzing stochastic policies. Combining the Bellman error-based analysis and tighter optimization equation, for $T$ interactions with the environment, we obtain a high-probability regret guarantee for objective which grows as $\\Tilde{O}(1/\\sqrt{T})$, excluding other factors. The proposed method can be applied for optimistic algorithms to obtain high-probability regret bounds and also be used for posterior sampling algorithms to obtain a loose Bayesian regret bounds but with significant improvement in computational complexity.", "authors": ["Mridul Agarwal", "Qinbo Bai", "Vaneet Aggarwal"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2021-09-12", "url": "https://arxiv.org/abs/2109.05439", "pdf_url": "https://arxiv.org/pdf/2109.05439v3", "arxiv_id": "2109.05439", "doi": null, "citation_count": 19, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "Transactions on Machine Learning Research, Dec 2022", "quality_score": 0.3253}
{"id": "b8550b3a313508224578a451d96ca8ffbdf7eb450d7d20f3c7760bc0029b35ef", "sources": ["arxiv", "semantic_scholar"], "title": "Quantum Continual Learning Overcoming Catastrophic Forgetting", "abstract": "Catastrophic forgetting describes the fact that machine learning models will likely forget the knowledge of previously learned tasks after the learning process of a new one. It is a vital problem in the continual learning scenario and recently has attracted tremendous concern across different communities. In this paper, we explore the catastrophic forgetting phenomena in the context of quantum machine learning. We find that, similar to those classical learning models based on neural networks, quantum learning systems likewise suffer from such forgetting problem in classification tasks emerging from various application scenes. We show that based on the local geometrical information in the loss function landscape of the trained model, a uniform strategy can be adapted to overcome the forgetting problem in the incremental learning setting. Our results uncover the catastrophic forgetting phenomena in quantum machine learning and offer a practical method to overcome this problem, which opens a new avenue for exploring potential quantum advantages towards continual learning.", "authors": ["Wenjie Jiang", "Zhide Lu", "Dong-Ling Deng"], "categories": ["cs.LG", "cond-mat.mes-hall", "quant-ph"], "fields_of_study": ["Physics", "Computer Science"], "published_date": "2021-08-05", "url": "https://arxiv.org/abs/2108.02786", "pdf_url": "https://arxiv.org/pdf/2108.02786v1", "arxiv_id": "2108.02786", "doi": "10.1088/0256-307X/39/5/050303", "citation_count": 13, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": "Chinese Physics Letters", "quality_score": 0.2865}
{"id": "c9054c2960a3bd170d8d9041ece091e8ab44bdefbf46c85f79bdccd03c29c14b", "sources": ["arxiv", "semantic_scholar"], "title": "A Pragmatic Look at Deep Imitation Learning", "abstract": "The introduction of the generative adversarial imitation learning (GAIL) algorithm has spurred the development of scalable imitation learning approaches using deep neural networks. Many of the algorithms that followed used a similar procedure, combining on-policy actor-critic algorithms with inverse reinforcement learning. More recently there have been an even larger breadth of approaches, most of which use off-policy algorithms. However, with the breadth of algorithms, everything from datasets to base reinforcement learning algorithms to evaluation settings can vary, making it difficult to fairly compare them. In this work we re-implement 6 different IL algorithms, updating 3 of them to be off-policy, base them on a common off-policy algorithm (SAC), and evaluate them on a widely-used expert trajectory dataset (D4RL) for the most common benchmark (MuJoCo). After giving all algorithms the same hyperparameter optimisation budget, we compare their results for a range of expert trajectories. In summary, GAIL, with all of its improvements, consistently performs well across a range of sample sizes, AdRIL is a simple contender that performs well with one important hyperparameter to tune, and behavioural cloning remains a strong baseline when data is more plentiful.", "authors": ["Kai Arulkumaran", "Dan Ogawa Lillrank"], "categories": ["cs.LG", "cs.NE", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2021-08-04", "url": "https://arxiv.org/abs/2108.01867", "pdf_url": "https://arxiv.org/pdf/2108.01867v2", "arxiv_id": "2108.01867", "doi": null, "citation_count": 13, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "Asian Conference on Machine Learning", "quality_score": 0.2865}
{"id": "e4fe80112dae1d20f6c5594c8971b7d2a20bc745e3c3ff486e7627b906132e29", "sources": ["arxiv", "semantic_scholar"], "title": "Open-Ended Learning Leads to Generally Capable Agents", "abstract": "In this work we create agents that can perform well beyond a single, individual task, that exhibit much wider generalisation of behaviour to a massive, rich space of challenges. We define a universe of tasks within an environment domain and demonstrate the ability to train agents that are generally capable across this vast space and beyond. The environment is natively multi-agent, spanning the continuum of competitive, cooperative, and independent games, which are situated within procedurally generated physical 3D worlds. The resulting space is exceptionally diverse in terms of the challenges posed to agents, and as such, even measuring the learning progress of an agent is an open research problem. We propose an iterative notion of improvement between successive generations of agents, rather than seeking to maximise a singular objective, allowing us to quantify progress despite tasks being incomparable in terms of achievable rewards. We show that through constructing an open-ended learning process, which dynamically changes the training task distributions and training objectives such that the agent never stops learning, we achieve consistent learning of new behaviours. The resulting agent is able to score reward in every one of our humanly solvable evaluation levels, with behaviour generalising to many held-out points in the universe of tasks. Examples of this zero-shot generalisation include good performance on Hide and Seek, Capture the Flag, and Tag. Through analysis and hand-authored probe tasks we characterise the behaviour of our agent, and find interesting emergent heuristic behaviours such as trial-and-error experimentation, simple tool use, option switching, and cooperation. Finally, we demonstrate that the general capabilities of this agent could unlock larger scale transfer of behaviour through cheap finetuning.", "authors": [" Open Ended Learning Team", "Adam Stooke", "Anuj Mahajan", "Catarina Barros", "Charlie Deck", "Jakob Bauer", "Jakub Sygnowski", "Maja Trebacz", "Max Jaderberg", "Michael Mathieu", "Nat McAleese", "Nathalie Bradley-Schmieg", "Nathaniel Wong", "Nicolas Porcel", "Roberta Raileanu", "Steph Hughes-Fitt", "Valentin Dalibard", "Wojciech Marian Czarnecki"], "categories": ["cs.LG", "cs.AI", "cs.MA"], "fields_of_study": ["Computer Science"], "published_date": "2021-07-27", "url": "https://arxiv.org/abs/2107.12808", "pdf_url": "https://arxiv.org/pdf/2107.12808v2", "arxiv_id": "2107.12808", "doi": null, "citation_count": 229, "influential_citation_count": 20, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.6611}
{"id": "5a585a2db2622fe24415f8e4c579a92c27323ef908b42a8535de8e977c7e2cae", "sources": ["arxiv", "semantic_scholar"], "title": "Continual Learning in the Teacher-Student Setup: Impact of Task Similarity", "abstract": "Continual learning-the ability to learn many tasks in sequence-is critical for artificial learning systems. Yet standard training methods for deep networks often suffer from catastrophic forgetting, where learning new tasks erases knowledge of earlier tasks. While catastrophic forgetting labels the problem, the theoretical reasons for interference between tasks remain unclear. Here, we attempt to narrow this gap between theory and practice by studying continual learning in the teacher-student setup. We extend previous analytical work on two-layer networks in the teacher-student setup to multiple teachers. Using each teacher to represent a different task, we investigate how the relationship between teachers affects the amount of forgetting and transfer exhibited by the student when the task switches. In line with recent work, we find that when tasks depend on similar features, intermediate task similarity leads to greatest forgetting. However, feature similarity is only one way in which tasks may be related. The teacher-student approach allows us to disentangle task similarity at the level of readouts (hidden-to-output weights) and features (input-to-hidden weights). We find a complex interplay between both types of similarity, initial transfer/forgetting rates, maximum transfer/forgetting, and long-term transfer/forgetting. Together, these results help illuminate the diverse factors contributing to catastrophic forgetting.", "authors": ["Sebastian Lee", "Sebastian Goldt", "Andrew Saxe"], "categories": ["stat.ML", "cond-mat.stat-mech", "cs.LG"], "fields_of_study": ["Computer Science", "Mathematics", "Physics"], "published_date": "2021-07-09", "url": "https://arxiv.org/abs/2107.04384", "pdf_url": "https://arxiv.org/pdf/2107.04384v1", "arxiv_id": "2107.04384", "doi": null, "citation_count": 101, "influential_citation_count": 8, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.5022}
{"id": "96831650473938045b2bb97bf70d143a7a2696506da667980858772ef2783209", "sources": ["arxiv", "semantic_scholar"], "title": "Asymptotics of Network Embeddings Learned via Subsampling", "abstract": "Network data are ubiquitous in modern machine learning, with tasks of interest including node classification, node clustering and link prediction. A frequent approach begins by learning an Euclidean embedding of the network, to which algorithms developed for vector-valued data are applied. For large networks, embeddings are learned using stochastic gradient methods where the sub-sampling scheme can be freely chosen. Despite the strong empirical performance of such methods, they are not well understood theoretically. Our work encapsulates representation methods using a subsampling approach, such as node2vec, into a single unifying framework. We prove, under the assumption that the graph is exchangeable, that the distribution of the learned embedding vectors asymptotically decouples. Moreover, we characterize the asymptotic distribution and provided rates of convergence, in terms of the latent parameters, which includes the choice of loss function and the embedding dimension. This provides a theoretical foundation to understand what the embedding vectors represent and how well these methods perform on downstream tasks. Notably, we observe that typically used loss functions may lead to shortcomings, such as a lack of Fisher consistency.", "authors": ["Andrew Davison", "Morgane Austern"], "categories": ["stat.ML", "cs.LG", "math.ST"], "fields_of_study": ["Mathematics", "Computer Science"], "published_date": "2021-07-06", "url": "https://arxiv.org/abs/2107.02363", "pdf_url": "https://arxiv.org/pdf/2107.02363v4", "arxiv_id": "2107.02363", "doi": null, "citation_count": 14, "influential_citation_count": 5, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3891}
{"id": "b9de67d24dcb926dd5996e0ececb23c39960df6f127bf6b57b8fc1e7128b786f", "sources": ["arxiv", "semantic_scholar"], "title": "Autoencoder based Randomized Learning of Feedforward Neural Networks for Regression", "abstract": "Feedforward neural networks are widely used as universal predictive models to fit data distribution. Common gradient-based learning, however, suffers from many drawbacks making the training process ineffective and time-consuming. Alternative randomized learning does not use gradients but selects hidden node parameters randomly. This makes the training process extremely fast. However, the problem in randomized learning is how to determine the random parameters. A recently proposed method uses autoencoders for unsupervised parameter learning. This method showed superior performance on classification tasks. In this work, we apply this method to regression problems, and, finding that it has some drawbacks, we show how to improve it. We propose a learning method of autoencoders that controls the produced random weights. We also propose how to determine the biases of hidden nodes. We empirically compare autoencoder based learning with other randomized learning methods proposed recently for regression and find that despite the proposed improvement of the autoencoder based learning, it does not outperform its competitors in fitting accuracy. Moreover, the method is much more complex than its competitors.", "authors": ["Grzegorz Dudek"], "categories": ["cs.LG", "cs.NE"], "fields_of_study": ["Computer Science"], "published_date": "2021-07-04", "url": "https://arxiv.org/abs/2107.01711", "pdf_url": "https://arxiv.org/pdf/2107.01711v1", "arxiv_id": "2107.01711", "doi": "10.1109/IJCNN52387.2021.9534263", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "IEEE International Joint Conference on Neural Network", "quality_score": 0.0753}
{"id": "50e26fff9b8c9bb47849fab7f5447a6797786f8fa0f474254db30b11d9c78b07", "sources": ["arxiv", "semantic_scholar"], "title": "Continual Competitive Memory: A Neural System for Online Task-Free Lifelong Learning", "abstract": "In this article, we propose a novel form of unsupervised learning, continual competitive memory (CCM), as well as a computational framework to unify related neural models that operate under the principles of competition. The resulting neural system is shown to offer an effective approach for combating catastrophic forgetting in online continual classification problems. We demonstrate that the proposed CCM system not only outperforms other competitive learning neural models but also yields performance that is competitive with several modern, state-of-the-art lifelong learning approaches on benchmarks such as Split MNIST and Split NotMNIST. CCM yields a promising path forward for acquiring representations that are robust to interference from data streams, especially when the task is unknown to the model and must be inferred without external guidance.", "authors": ["Alexander G. Ororbia"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2021-06-24", "url": "https://arxiv.org/abs/2106.13300", "pdf_url": "https://arxiv.org/pdf/2106.13300v1", "arxiv_id": "2106.13300", "doi": "10.31219/osf.io/yw6ua", "citation_count": 7, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2258}
{"id": "d4364f12085a5faa51a1560f84b3b901044a2443217e2aeea8fb47f4a170d022", "sources": ["arxiv", "semantic_scholar"], "title": "Position-Sensing Graph Neural Networks: Proactively Learning Nodes Relative Positions", "abstract": "Most existing graph neural networks (GNNs) learn node embeddings using the framework of message passing and aggregation. Such GNNs are incapable of learning relative positions between graph nodes within a graph. To empower GNNs with the awareness of node positions, some nodes are set as anchors. Then, using the distances from a node to the anchors, GNNs can infer relative positions between nodes. However, P-GNNs arbitrarily select anchors, leading to compromising position-awareness and feature extraction. To eliminate this compromise, we demonstrate that selecting evenly distributed and asymmetric anchors is essential. On the other hand, we show that choosing anchors that can aggregate embeddings of all the nodes within a graph is NP-complete. Therefore, devising efficient optimal algorithms in a deterministic approach is practically not feasible. To ensure position-awareness and bypass NP-completeness, we propose Position-Sensing Graph Neural Networks (PSGNNs), learning how to choose anchors in a back-propagatable fashion. Experiments verify the effectiveness of PSGNNs against state-of-the-art GNNs, substantially improving performance on various synthetic and real-world graph datasets while enjoying stable scalability. Specifically, PSGNNs on average boost AUC more than 14% for pairwise node classification and 18% for link prediction over the existing state-of-the-art position-aware methods. Our source code is publicly available at: https://github.com/ZhenyueQin/PSGNN.", "authors": ["Zhenyue Qin", "Yiqun Zhang Saeed Anwar", "Dongwoo Kim", "Yang Liu", "Pan Ji", "Tom Gedeon"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science", "Medicine"], "published_date": "2021-05-24", "url": "https://arxiv.org/abs/2105.11346", "pdf_url": "https://arxiv.org/pdf/2105.11346v2", "arxiv_id": "2105.11346", "doi": "10.1109/TNNLS.2024.3374464", "citation_count": 2, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/ZhenyueQin/PSGNN", "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.1193}
{"id": "6a13a14848f5b7550fef3e4cbc75f1ba1b998adb328b15835b68994c65e2216b", "sources": ["arxiv", "semantic_scholar"], "title": "Feature Encoding with AutoEncoders for Weakly-supervised Anomaly Detection", "abstract": "Weakly-supervised anomaly detection aims at learning an anomaly detector from a limited amount of labeled data and abundant unlabeled data. Recent works build deep neural networks for anomaly detection by discriminatively mapping the normal samples and abnormal samples to different regions in the feature space or fitting different distributions. However, due to the limited number of annotated anomaly samples, directly training networks with the discriminative loss may not be sufficient. To overcome this issue, this paper proposes a novel strategy to transform the input data into a more meaningful representation that could be used for anomaly detection. Specifically, we leverage an autoencoder to encode the input data and utilize three factors, hidden representation, reconstruction residual vector, and reconstruction error, as the new representation for the input data. This representation amounts to encode a test sample with its projection on the training data manifold, its direction to its projection and its distance to its projection. In addition to this encoding, we also propose a novel network architecture to seamlessly incorporate those three factors. From our extensive experiments, the benefits of the proposed strategy are clearly demonstrated by its superior performance over the competitive methods.", "authors": ["Yingjie Zhou", "Xucheng Song", "Yanru Zhang", "Fanxing Liu", "Ce Zhu", "Lingqiao Liu"], "categories": ["cs.LG", "cs.NI"], "fields_of_study": ["Computer Science", "Medicine"], "published_date": "2021-05-22", "url": "https://arxiv.org/abs/2105.10500", "pdf_url": "https://arxiv.org/pdf/2105.10500v3", "arxiv_id": "2105.10500", "doi": "10.1109/TNNLS.2021.3086137", "citation_count": 150, "influential_citation_count": 22, "has_code": false, "code_url": null, "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.6809}
{"id": "65d13e6c4d80606088db0b8158b0f3783869b858e2c72d8239288f77b76651fc", "sources": ["arxiv", "semantic_scholar"], "title": "Statistical Mechanical Analysis of Catastrophic Forgetting in Continual Learning with Teacher and Student Networks", "abstract": "When a computational system continuously learns from an ever-changing environment, it rapidly forgets its past experiences. This phenomenon is called catastrophic forgetting. While a line of studies has been proposed with respect to avoiding catastrophic forgetting, most of the methods are based on intuitive insights into the phenomenon, and their performances have been evaluated by numerical experiments using benchmark datasets. Therefore, in this study, we provide the theoretical framework for analyzing catastrophic forgetting by using teacher-student learning. Teacher-student learning is a framework in which we introduce two neural networks: one neural network is a target function in supervised learning, and the other is a learning neural network. To analyze continual learning in the teacher-student framework, we introduce the similarity of the input distribution and the input-output relationship of the target functions as the similarity of tasks. In this theoretical framework, we also provide a qualitative understanding of how a single-layer linear learning neural network forgets tasks. Based on the analysis, we find that the network can avoid catastrophic forgetting when the similarity among input distributions is small and that of the input-output relationship of the target functions is large. The analysis also suggests that a system often exhibits a characteristic phenomenon called overshoot, which means that even if the learning network has once undergone catastrophic forgetting, it is possible that the network may perform reasonably well after further learning of the current task.", "authors": ["Haruka Asanuma", "Shiro Takagi", "Yoshihiro Nagano", "Yuki Yoshida", "Yasuhiko Igarashi", "Masato Okada"], "categories": ["stat.ML", "cs.LG"], "fields_of_study": ["Mathematics", "Computer Science"], "published_date": "2021-05-16", "url": "https://arxiv.org/abs/2105.07385", "pdf_url": "https://arxiv.org/pdf/2105.07385v1", "arxiv_id": "2105.07385", "doi": "10.7566/JPSJ.90.104001", "citation_count": 25, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": "Journal of the Physical Society of Japan", "quality_score": 0.3537}
{"id": "18c813deeb0c7e87652b14f7bb33405205f7f27e2c6517e52a44a3768f043854", "sources": ["arxiv", "semantic_scholar"], "title": "TAG: Task-based Accumulated Gradients for Lifelong learning", "abstract": "When an agent encounters a continual stream of new tasks in the lifelong learning setting, it leverages the knowledge it gained from the earlier tasks to help learn the new tasks better. In such a scenario, identifying an efficient knowledge representation becomes a challenging problem. Most research works propose to either store a subset of examples from the past tasks in a replay buffer, dedicate a separate set of parameters to each task or penalize excessive updates over parameters by introducing a regularization term. While existing methods employ the general task-agnostic stochastic gradient descent update rule, we propose a task-aware optimizer that adapts the learning rate based on the relatedness among tasks. We utilize the directions taken by the parameters during the updates by accumulating the gradients specific to each task. These task-based accumulated gradients act as a knowledge base that is maintained and updated throughout the stream. We empirically show that our proposed adaptive learning rate not only accounts for catastrophic forgetting but also allows positive backward transfer. We also show that our method performs better than several state-of-the-art methods in lifelong learning on complex datasets with a large number of tasks.", "authors": ["Pranshu Malviya", "Balaraman Ravindran", "Sarath Chandar"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2021-05-11", "url": "https://arxiv.org/abs/2105.05155", "pdf_url": "https://arxiv.org/pdf/2105.05155v3", "arxiv_id": "2105.05155", "doi": null, "citation_count": 8, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2386}
{"id": "0b7a74ce55c275b9c3b2462d1c4eb3776b1640577e9134b046076442bc7fda11", "sources": ["arxiv", "semantic_scholar"], "title": "The Modern Mathematics of Deep Learning", "abstract": "We describe the new field of mathematical analysis of deep learning. This field emerged around a list of research questions that were not answered within the classical framework of learning theory. These questions concern: the outstanding generalization power of overparametrized neural networks, the role of depth in deep architectures, the apparent absence of the curse of dimensionality, the surprisingly successful optimization performance despite the non-convexity of the problem, understanding what features are learned, why deep architectures perform exceptionally well in physical problems, and which fine aspects of an architecture affect the behavior of a learning task in which way. We present an overview of modern approaches that yield partial answers to these questions. For selected approaches, we describe the main ideas in more detail.", "authors": ["Julius Berner", "Philipp Grohs", "Gitta Kutyniok", "Philipp Petersen"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2021-05-09", "url": "https://arxiv.org/abs/2105.04026", "pdf_url": "https://arxiv.org/pdf/2105.04026v2", "arxiv_id": "2105.04026", "doi": "10.1017/9781009025096.002", "citation_count": 134, "influential_citation_count": 8, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.5326}
{"id": "493262ba23a2233010fcc66ebdec692bcb68760006b87efda2ea4d692825f166", "sources": ["arxiv", "semantic_scholar"], "title": "Noether's Learning Dynamics: Role of Symmetry Breaking in Neural Networks", "abstract": "In nature, symmetry governs regularities, while symmetry breaking brings texture. In artificial neural networks, symmetry has been a central design principle to efficiently capture regularities in the world, but the role of symmetry breaking is not well understood. Here, we develop a theoretical framework to study the \"geometry of learning dynamics\" in neural networks, and reveal a key mechanism of explicit symmetry breaking behind the efficiency and stability of modern neural networks. To build this understanding, we model the discrete learning dynamics of gradient descent using a continuous-time Lagrangian formulation, in which the learning rule corresponds to the kinetic energy and the loss function corresponds to the potential energy. Then, we identify \"kinetic symmetry breaking\" (KSB), the condition when the kinetic energy explicitly breaks the symmetry of the potential function. We generalize Noether's theorem known in physics to take into account KSB and derive the resulting motion of the Noether charge: \"Noether's Learning Dynamics\" (NLD). Finally, we apply NLD to neural networks with normalization layers and reveal how KSB introduces a mechanism of \"implicit adaptive optimization\", establishing an analogy between learning dynamics induced by normalization layers and RMSProp. Overall, through the lens of Lagrangian mechanics, we have established a theoretical foundation to discover geometric design principles for the learning dynamics of neural networks.", "authors": ["Hidenori Tanaka", "Daniel Kunin"], "categories": ["cs.LG", "cond-mat.dis-nn", "cond-mat.stat-mech", "q-bio.NC", "stat.ML"], "fields_of_study": ["Computer Science", "Physics", "Biology", "Mathematics"], "published_date": "2021-05-06", "url": "https://arxiv.org/abs/2105.02716", "pdf_url": "https://arxiv.org/pdf/2105.02716v2", "arxiv_id": "2105.02716", "doi": null, "citation_count": 50, "influential_citation_count": 6, "has_code": false, "code_url": null, "venue": "Neural Information Processing Systems", "quality_score": 0.4269}
{"id": "6349c1f85dca1e78446dfb9d249fead664a46ed09a42dd0711313950631a9b9b", "sources": ["arxiv", "semantic_scholar"], "title": "Continual Distributed Learning for Crisis Management", "abstract": "Social media platforms such as Twitter, Facebook etc can be utilised as an important source of information during disaster events. This information can be used for disaster response and crisis management if processed accurately and quickly. However, the data present in such situations is ever-changing, and using considerable resources during such a crisis is not feasible. Therefore, we have to develop a low resource and continually learning system that incorporates text classification models which are robust against noisy and unordered data. We utilised Distributed learning which enabled us to learn on resource-constrained devices, then to alleviate catastrophic forgetting in our target neural networks we utilized regularization. We then applied federated averaging for distributed learning and to aggregate the central model for continual learning.", "authors": ["Aman Priyanshu", "Mudit Sinha", "Shreyans Mehta"], "categories": ["cs.LG", "cs.DC"], "fields_of_study": ["Computer Science"], "published_date": "2021-04-26", "url": "https://arxiv.org/abs/2104.12876", "pdf_url": "https://arxiv.org/pdf/2104.12876v2", "arxiv_id": "2104.12876", "doi": null, "citation_count": 5, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1945}
{"id": "c0b39f14661b63a42b0d49c31cfd39b68e69f476b4aa8154d629ebffd723a830", "sources": ["arxiv", "semantic_scholar"], "title": "Class-Incremental Learning with Generative Classifiers", "abstract": "Incrementally training deep neural networks to recognize new classes is a challenging problem. Most existing class-incremental learning methods store data or use generative replay, both of which have drawbacks, while 'rehearsal-free' alternatives such as parameter regularization or bias-correction methods do not consistently achieve high performance. Here, we put forward a new strategy for class-incremental learning: generative classification. Rather than directly learning the conditional distribution p(y|x), our proposal is to learn the joint distribution p(x,y), factorized as p(x|y)p(y), and to perform classification using Bayes' rule. As a proof-of-principle, here we implement this strategy by training a variational autoencoder for each class to be learned and by using importance sampling to estimate the likelihoods p(x|y). This simple approach performs very well on a diverse set of continual learning benchmarks, outperforming generative replay and other existing baselines that do not store data.", "authors": ["Gido M. van de Ven", "Zhe Li", "Andreas S. Tolias"], "categories": ["cs.LG", "cs.AI", "cs.CV", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2021-04-20", "url": "https://arxiv.org/abs/2104.10093", "pdf_url": "https://arxiv.org/pdf/2104.10093v2", "arxiv_id": "2104.10093", "doi": "10.1109/CVPRW53098.2021.00400", "citation_count": 76, "influential_citation_count": 5, "has_code": false, "code_url": null, "venue": "Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2021, pp. 3611-3620", "quality_score": 0.4716}
{"id": "7fd685d28f9f6768fb2fbc7205907ab54ca1b7405ad9cc2767087dede0c573f0", "sources": ["arxiv", "semantic_scholar"], "title": "Neural Architecture Search of Deep Priors: Towards Continual Learning without Catastrophic Interference", "abstract": "In this paper we analyze the classification performance of neural network structures without parametric inference. Making use of neural architecture search, we empirically demonstrate that it is possible to find random weight architectures, a deep prior, that enables a linear classification to perform on par with fully trained deep counterparts. Through ablation experiments, we exclude the possibility of winning a weight initialization lottery and confirm that suitable deep priors do not require additional inference. In an extension to continual learning, we investigate the possibility of catastrophic interference free incremental learning. Under the assumption of classes originating from the same data distribution, a deep prior found on only a subset of classes is shown to allow discrimination of further classes through training of a simple linear classifier.", "authors": ["Martin Mundt", "Iuliia Pliushch", "Visvanathan Ramesh"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2021-04-14", "url": "https://arxiv.org/abs/2104.06788", "pdf_url": "https://arxiv.org/pdf/2104.06788v1", "arxiv_id": "2104.06788", "doi": "10.1109/CVPRW53098.2021.00391", "citation_count": 8, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2386}
{"id": "55ac9b9b66e071fa4a8319ca7bb31a8e8d94bdfb68eb9855a64bc6500079c1c8", "sources": ["arxiv", "semantic_scholar"], "title": "FedGraphNN: A Federated Learning System and Benchmark for Graph Neural Networks", "abstract": "Graph Neural Network (GNN) research is rapidly growing thanks to the capacity of GNNs in learning distributed representations from graph-structured data. However, centralizing a massive amount of real-world graph data for GNN training is prohibitive due to privacy concerns, regulation restrictions, and commercial competitions. Federated learning (FL), a trending distributed learning paradigm, provides possibilities to solve this challenge while preserving data privacy. Despite recent advances in vision and language domains, there is no suitable platform for the FL of GNNs. To this end, we introduce FedGraphNN, an open FL benchmark system that can facilitate research on federated GNNs. FedGraphNN is built on a unified formulation of graph FL and contains a wide range of datasets from different domains, popular GNN models, and FL algorithms, with secure and efficient system support. Particularly for the datasets, we collect, preprocess, and partition 36 datasets from 7 domains, including both publicly available ones and specifically obtained ones such as hERG and Tencent. Our empirical analysis showcases the utility of our benchmark system, while exposing significant challenges in graph FL: federated GNNs perform worse in most datasets with a non-IID split than centralized GNNs; the GNN model that attains the best result in the centralized setting may not maintain its advantage in the FL setting. These results imply that more research efforts are needed to unravel the mystery behind federated GNNs. Moreover, our system performance analysis demonstrates that the FedGraphNN system is computationally efficient and secure to large-scale graphs datasets. We maintain the source code at https://github.com/FedML-AI/FedGraphNN.", "authors": ["Chaoyang He", "Keshav Balasubramanian", "Emir Ceyani", "Carl Yang", "Han Xie", "Lichao Sun", "Lifang He", "Liangwei Yang", "Philip S. Yu", "Yu Rong", "Peilin Zhao", "Junzhou Huang", "Murali Annavaram", "Salman Avestimehr"], "categories": ["cs.LG", "cs.AI", "cs.DC"], "fields_of_study": ["Computer Science"], "published_date": "2021-04-14", "url": "https://arxiv.org/abs/2104.07145", "pdf_url": "https://arxiv.org/pdf/2104.07145v2", "arxiv_id": "2104.07145", "doi": null, "citation_count": 2, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/FedML-AI/FedGraphNN", "venue": null, "quality_score": 0.1193}
{"id": "9c11cd8234e09e1a3636b413a2c41cf57bf1fabe986fc44b01fdf3f1276f2d60", "sources": ["arxiv", "semantic_scholar"], "title": "Towards Lifelong Learning of End-to-end ASR", "abstract": "Automatic speech recognition (ASR) technologies today are primarily optimized for given datasets; thus, any changes in the application environment (e.g., acoustic conditions or topic domains) may inevitably degrade the performance. We can collect new data describing the new environment and fine-tune the system, but this naturally leads to higher error rates for the earlier datasets, referred to as catastrophic forgetting. The concept of lifelong learning (LLL) aiming to enable a machine to sequentially learn new tasks from new datasets describing the changing real world without forgetting the previously learned knowledge is thus brought to attention. This paper reports, to our knowledge, the first effort to extensively consider and analyze the use of various approaches of LLL in end-to-end (E2E) ASR, including proposing novel methods in saving data for past domains to mitigate the catastrophic forgetting problem. An overall relative reduction of 28.7% in WER was achieved compared to the fine-tuning baseline when sequentially learning on three very different benchmark corpora. This can be the first step toward the highly desired ASR technologies capable of synchronizing with the continuously changing real world.", "authors": ["Heng-Jui Chang", "Hung-yi Lee", "Lin-shan Lee"], "categories": ["cs.CL", "eess.AS"], "fields_of_study": ["Computer Science", "Engineering"], "published_date": "2021-04-04", "url": "https://arxiv.org/abs/2104.01616", "pdf_url": "https://arxiv.org/pdf/2104.01616v3", "arxiv_id": "2104.01616", "doi": "10.21437/interspeech.2021-563", "citation_count": 40, "influential_citation_count": 4, "has_code": false, "code_url": null, "venue": "Interspeech", "quality_score": 0.4032}
{"id": "dde51736d56d0d5121be26b086371fc46e5a116299574a5669f7c165e086d860", "sources": ["arxiv", "semantic_scholar"], "title": "Supervised Contrastive Replay: Revisiting the Nearest Class Mean Classifier in Online Class-Incremental Continual Learning", "abstract": "Online class-incremental continual learning (CL) studies the problem of learning new classes continually from an online non-stationary data stream, intending to adapt to new data while mitigating catastrophic forgetting. While memory replay has shown promising results, the recency bias in online learning caused by the commonly used Softmax classifier remains an unsolved challenge. Although the Nearest-Class-Mean (NCM) classifier is significantly undervalued in the CL community, we demonstrate that it is a simple yet effective substitute for the Softmax classifier. It addresses the recency bias and avoids structural changes in the fully-connected layer for new classes. Moreover, we observe considerable and consistent performance gains when replacing the Softmax classifier with the NCM classifier for several state-of-the-art replay methods. To leverage the NCM classifier more effectively, data embeddings belonging to the same class should be clustered and well-separated from those with a different class label. To this end, we contribute Supervised Contrastive Replay (SCR), which explicitly encourages samples from the same class to cluster tightly in embedding space while pushing those of different classes further apart during replay-based training. Overall, we observe that our proposed SCR substantially reduces catastrophic forgetting and outperforms state-of-the-art CL methods by a significant margin on a variety of datasets.", "authors": ["Zheda Mai", "Ruiwen Li", "Hyunwoo Kim", "Scott Sanner"], "categories": ["cs.LG", "cs.AI", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2021-03-22", "url": "https://arxiv.org/abs/2103.13885", "pdf_url": "https://arxiv.org/pdf/2103.13885v3", "arxiv_id": "2103.13885", "doi": "10.1109/CVPRW53098.2021.00398", "citation_count": 234, "influential_citation_count": 46, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.836}
{"id": "4491f0c310151840e7f5761a9241bcd24352159229e872b25ee31846cac0f7a1", "sources": ["arxiv", "semantic_scholar"], "title": "Catastrophic Forgetting in Deep Graph Networks: an Introductory Benchmark for Graph Classification", "abstract": "In this work, we study the phenomenon of catastrophic forgetting in the graph representation learning scenario. The primary objective of the analysis is to understand whether classical continual learning techniques for flat and sequential data have a tangible impact on performances when applied to graph data. To do so, we experiment with a structure-agnostic model and a deep graph network in a robust and controlled environment on three different datasets. The benchmark is complemented by an investigation on the effect of structure-preserving regularization techniques on catastrophic forgetting. We find that replay is the most effective strategy in so far, which also benefits the most from the use of regularization. Our findings suggest interesting future research at the intersection of the continual and graph representation learning fields. Finally, we provide researchers with a flexible software framework to reproduce our results and carry out further experiments.", "authors": ["Antonio Carta", "Andrea Cossu", "Federico Errica", "Davide Bacciu"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2021-03-22", "url": "https://arxiv.org/abs/2103.11750", "pdf_url": "https://arxiv.org/pdf/2103.11750v1", "arxiv_id": "2103.11750", "doi": null, "citation_count": 20, "influential_citation_count": 1, "has_code": true, "code_url": "https://github.com/diningphil/continual_learning_for_graphs", "venue": "arXiv.org", "quality_score": 0.3306}
{"id": "3893663afda3168eb4047e81d6d9c6bc25b721dd4c67d50549903de416f27f0d", "sources": ["arxiv", "semantic_scholar"], "title": "Gradient Projection Memory for Continual Learning", "abstract": "The ability to learn continually without forgetting the past tasks is a desired attribute for artificial learning systems. Existing approaches to enable such learning in artificial neural networks usually rely on network growth, importance based weight update or replay of old data from the memory. In contrast, we propose a novel approach where a neural network learns new tasks by taking gradient steps in the orthogonal direction to the gradient subspaces deemed important for the past tasks. We find the bases of these subspaces by analyzing network representations (activations) after learning each task with Singular Value Decomposition (SVD) in a single shot manner and store them in the memory as Gradient Projection Memory (GPM). With qualitative and quantitative analyses, we show that such orthogonal gradient descent induces minimum to no interference with the past tasks, thereby mitigates forgetting. We evaluate our algorithm on diverse image classification datasets with short and long sequences of tasks and report better or on-par performance compared to the state-of-the-art approaches.", "authors": ["Gobinda Saha", "Isha Garg", "Kaushik Roy"], "categories": ["cs.LG", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2021-03-17", "url": "https://arxiv.org/abs/2103.09762", "pdf_url": "https://arxiv.org/pdf/2103.09762v1", "arxiv_id": "2103.09762", "doi": null, "citation_count": 439, "influential_citation_count": 83, "has_code": false, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.9621}
{"id": "3e6e194ab04d772f7c9a4b361d812ce0f755d0c92dfa14b49eea8c1f8c98ae8b", "sources": ["arxiv", "semantic_scholar"], "title": "Continual Learning for Recurrent Neural Networks: an Empirical Evaluation", "abstract": "Learning continuously during all model lifetime is fundamental to deploy machine learning solutions robust to drifts in the data distribution. Advances in Continual Learning (CL) with recurrent neural networks could pave the way to a large number of applications where incoming data is non stationary, like natural language processing and robotics. However, the existing body of work on the topic is still fragmented, with approaches which are application-specific and whose assessment is based on heterogeneous learning protocols and datasets. In this paper, we organize the literature on CL for sequential data processing by providing a categorization of the contributions and a review of the benchmarks. We propose two new benchmarks for CL with sequential data based on existing datasets, whose characteristics resemble real-world applications. We also provide a broad empirical evaluation of CL and Recurrent Neural Networks in class-incremental scenario, by testing their ability to mitigate forgetting with a number of different strategies which are not specific to sequential data processing. Our results highlight the key role played by the sequence length and the importance of a clear specification of the CL scenario.", "authors": ["Andrea Cossu", "Antonio Carta", "Vincenzo Lomonaco", "Davide Bacciu"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Medicine", "Computer Science"], "published_date": "2021-03-12", "url": "https://arxiv.org/abs/2103.07492", "pdf_url": "https://arxiv.org/pdf/2103.07492v4", "arxiv_id": "2103.07492", "doi": "10.1016/j.neunet.2021.07.021", "citation_count": 125, "influential_citation_count": 5, "has_code": false, "code_url": null, "venue": "Neural Networks", "quality_score": 0.5251}
{"id": "49ee26fdf5ce1cb708a1a8d9f9cbe68cc512340a6bb08a30388741bcb790bb70", "sources": ["arxiv", "semantic_scholar"], "title": "Selective Replay Enhances Learning in Online Continual Analogical Reasoning", "abstract": "In continual learning, a system learns from non-stationary data streams or batches without catastrophic forgetting. While this problem has been heavily studied in supervised image classification and reinforcement learning, continual learning in neural networks designed for abstract reasoning has not yet been studied. Here, we study continual learning of analogical reasoning. Analogical reasoning tests such as Raven's Progressive Matrices (RPMs) are commonly used to measure non-verbal abstract reasoning in humans, and recently offline neural networks for the RPM problem have been proposed. In this paper, we establish experimental baselines, protocols, and forward and backward transfer metrics to evaluate continual learners on RPMs. We employ experience replay to mitigate catastrophic forgetting. Prior work using replay for image classification tasks has found that selectively choosing the samples to replay offers little, if any, benefit over random selection. In contrast, we find that selective replay can significantly outperform random selection for the RPM task.", "authors": ["Tyler L. Hayes", "Christopher Kanan"], "categories": ["cs.AI", "cs.CV", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2021-03-06", "url": "https://arxiv.org/abs/2103.03987", "pdf_url": "https://arxiv.org/pdf/2103.03987v2", "arxiv_id": "2103.03987", "doi": "10.1109/CVPRW53098.2021.00389", "citation_count": 26, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.3578}
{"id": "9b54446bf32a252279c0920a4d83acb2b8805b3c53ed3419c6cc121cea8747ce", "sources": ["arxiv", "semantic_scholar"], "title": "Learning to Continually Learn Rapidly from Few and Noisy Data", "abstract": "Neural networks suffer from catastrophic forgetting and are unable to sequentially learn new tasks without guaranteed stationarity in data distribution. Continual learning could be achieved via replay -- by concurrently training externally stored old data while learning a new task. However, replay becomes less effective when each past task is allocated with less memory. To overcome this difficulty, we supplemented replay mechanics with meta-learning for rapid knowledge acquisition. By employing a meta-learner, which \\textit{learns a learning rate per parameter per past task}, we found that base learners produced strong results when less memory was available. Additionally, our approach inherited several meta-learning advantages for continual learning: it demonstrated strong robustness to continually learn under the presence of noises and yielded base learners to higher accuracy in less updates.", "authors": ["Nicholas I-Hsien Kuo", "Mehrtash Harandi", "Nicolas Fourrier", "Christian Walder", "Gabriela Ferraro", "Hanna Suominen"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2021-03-06", "url": "https://arxiv.org/abs/2103.04066", "pdf_url": "https://arxiv.org/pdf/2103.04066v1", "arxiv_id": "2103.04066", "doi": null, "citation_count": 4, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.1747}
{"id": "17e943dbee24d2e9baeaaf4eac66284ffc67248dd91721b2d2a1f8cd5eb067c5", "sources": ["arxiv", "semantic_scholar"], "title": "Continuous Coordination As a Realistic Scenario for Lifelong Learning", "abstract": "Current deep reinforcement learning (RL) algorithms are still highly task-specific and lack the ability to generalize to new environments. Lifelong learning (LLL), however, aims at solving multiple tasks sequentially by efficiently transferring and using knowledge between tasks. Despite a surge of interest in lifelong RL in recent years, the lack of a realistic testbed makes robust evaluation of LLL algorithms difficult. Multi-agent RL (MARL), on the other hand, can be seen as a natural scenario for lifelong RL due to its inherent non-stationarity, since the agents' policies change over time. In this work, we introduce a multi-agent lifelong learning testbed that supports both zero-shot and few-shot settings. Our setup is based on Hanabi -- a partially-observable, fully cooperative multi-agent game that has been shown to be challenging for zero-shot coordination. Its large strategy space makes it a desirable environment for lifelong RL tasks. We evaluate several recent MARL methods, and benchmark state-of-the-art LLL algorithms in limited memory and computation regimes to shed light on their strengths and weaknesses. This continual learning paradigm also provides us with a pragmatic way of going beyond centralized training which is the most commonly used training protocol in MARL. We empirically show that the agents trained in our setup are able to coordinate well with unseen agents, without any additional assumptions made by previous works. The code and all pre-trained models are available at https://github.com/chandar-lab/Lifelong-Hanabi.", "authors": ["Hadi Nekoei", "Akilesh Badrinaaraayanan", "Aaron Courville", "Sarath Chandar"], "categories": ["cs.LG", "cs.AI", "cs.MA"], "fields_of_study": ["Computer Science"], "published_date": "2021-03-04", "url": "https://arxiv.org/abs/2103.03216", "pdf_url": "https://arxiv.org/pdf/2103.03216v2", "arxiv_id": "2103.03216", "doi": null, "citation_count": 51, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/chandar-lab/Lifelong-Hanabi", "venue": "International Conference on Machine Learning", "quality_score": 0.429}
{"id": "3b171e001ccacc2bcb3d5b73d9f01d8823e5f7a3da6a00b7ed26bc9c10b96eaa", "sources": ["arxiv", "semantic_scholar"], "title": "Significance tests of feature relevance for a black-box learner", "abstract": "An exciting recent development is the uptake of deep neural networks in many scientific fields, where the main objective is outcome prediction with the black-box nature. Significance testing is promising to address the black-box issue and explore novel scientific insights and interpretation of the decision-making process based on a deep learning model. However, testing for a neural network poses a challenge because of its black-box nature and unknown limiting distributions of parameter estimates while existing methods require strong assumptions or excessive computation. In this article, we derive one-split and two-split tests relaxing the assumptions and computational complexity of existing black-box tests and extending to examine the significance of a collection of features of interest in a dataset of possibly a complex type such as an image. The one-split test estimates and evaluates a black-box model based on estimation and inference subsets through sample splitting and data perturbation. The two-split test further splits the inference subset into two but require no perturbation. Also, we develop their combined versions by aggregating the p-values based on repeated sample splitting. By deflating the bias-sd-ratio, we establish asymptotic null distributions of the test statistics and the consistency in terms of Type II error. Numerically, we demonstrate the utility of the proposed tests on seven simulated examples and six real datasets. Accompanying this paper is our Python library dnn-inference (https://dnn-inference.readthedocs.io/en/latest/) that implements the proposed tests.", "authors": ["Ben Dai", "Xiaotong Shen", "Wei Pan"], "categories": ["stat.ML", "cs.LG", "stat.ME"], "fields_of_study": ["Medicine", "Computer Science", "Mathematics"], "published_date": "2021-03-02", "url": "https://arxiv.org/abs/2103.04985", "pdf_url": "https://arxiv.org/pdf/2103.04985v3", "arxiv_id": "2103.04985", "doi": "10.1109/TNNLS.2022.3185742", "citation_count": 39, "influential_citation_count": 6, "has_code": false, "code_url": null, "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.4225}
{"id": "141ba209d9fed665f6ada9baef1fbc5f869d417c31888aa8101cad64827fca20", "sources": ["arxiv", "semantic_scholar"], "title": "Anomaly Detection on Attributed Networks via Contrastive Self-Supervised Learning", "abstract": "Anomaly detection on attributed networks attracts considerable research interests due to wide applications of attributed networks in modeling a wide range of complex systems. Recently, the deep learning-based anomaly detection methods have shown promising results over shallow approaches, especially on networks with high-dimensional attributes and complex structures. However, existing approaches, which employ graph autoencoder as their backbone, do not fully exploit the rich information of the network, resulting in suboptimal performance. Furthermore, these methods do not directly target anomaly detection in their learning objective and fail to scale to large networks due to the full graph training mechanism. To overcome these limitations, in this paper, we present a novel contrastive self-supervised learning framework for anomaly detection on attributed networks. Our framework fully exploits the local information from network data by sampling a novel type of contrastive instance pair, which can capture the relationship between each node and its neighboring substructure in an unsupervised way. Meanwhile, a well-designed graph neural network-based contrastive learning model is proposed to learn informative embedding from high-dimensional attributes and local structure and measure the agreement of each instance pairs with its outputted scores. The multi-round predicted scores by the contrastive learning model are further used to evaluate the abnormality of each node with statistical estimation. In this way, the learning model is trained by a specific anomaly detection-aware target. Furthermore, since the input of the graph neural network module is batches of instance pairs instead of the full network, our framework can adapt to large networks flexibly. Experimental results show that our proposed framework outperforms the state-of-the-art baseline methods on all seven benchmark datasets.", "authors": ["Yixin Liu", "Zhao Li", "Shirui Pan", "Chen Gong", "Chuan Zhou", "George Karypis"], "categories": ["cs.LG"], "fields_of_study": ["Medicine", "Computer Science"], "published_date": "2021-02-27", "url": "https://arxiv.org/abs/2103.00113", "pdf_url": "https://arxiv.org/pdf/2103.00113v2", "arxiv_id": "2103.00113", "doi": "10.1109/TNNLS.2021.3068344", "citation_count": 475, "influential_citation_count": 71, "has_code": false, "code_url": null, "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.9287}
{"id": "94cfe88f95a09aa9a76cf5ce18dd1d1ead579597464cc2c685e228004d8db172", "sources": ["arxiv", "semantic_scholar"], "title": "Understanding Catastrophic Forgetting and Remembering in Continual Learning with Optimal Relevance Mapping", "abstract": "Catastrophic forgetting in neural networks is a significant problem for continual learning. A majority of the current methods replay previous data during training, which violates the constraints of an ideal continual learning system. Additionally, current approaches that deal with forgetting ignore the problem of catastrophic remembering, i.e. the worsening ability to discriminate between data from different tasks. In our work, we introduce Relevance Mapping Networks (RMNs) which are inspired by the Optimal Overlap Hypothesis. The mappings reflects the relevance of the weights for the task at hand by assigning large weights to essential parameters. We show that RMNs learn an optimized representational overlap that overcomes the twin problem of catastrophic forgetting and remembering. Our approach achieves state-of-the-art performance across all common continual learning datasets, even significantly outperforming data replay methods while not violating the constraints for an ideal continual learning system. Moreover, RMNs retain the ability to detect data from new tasks in an unsupervised manner, thus proving their resilience against catastrophic remembering.", "authors": ["Prakhar Kaushik", "Alex Gain", "Adam Kortylewski", "Alan Yuille"], "categories": ["cs.LG", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2021-02-22", "url": "https://arxiv.org/abs/2102.11343", "pdf_url": "https://arxiv.org/pdf/2102.11343v1", "arxiv_id": "2102.11343", "doi": null, "citation_count": 79, "influential_citation_count": 11, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.5396}
{"id": "4254fdc4f52fcfd31e1846cdf9efb4472da4e76574381e9d8e932d5cdf564935", "sources": ["arxiv", "semantic_scholar"], "title": "Does the Adam Optimizer Exacerbate Catastrophic Forgetting?", "abstract": "Catastrophic forgetting remains a severe hindrance to the broad application of artificial neural networks (ANNs), however, it continues to be a poorly understood phenomenon. Despite the extensive amount of work on catastrophic forgetting, we argue that it is still unclear how exactly the phenomenon should be quantified, and, moreover, to what degree all of the choices we make when designing learning systems affect the amount of catastrophic forgetting. We use various testbeds from the reinforcement learning and supervised learning literature to (1) provide evidence that the choice of which modern gradient-based optimization algorithm is used to train an ANN has a significant impact on the amount of catastrophic forgetting and show that-surprisingly-in many instances classical algorithms such as vanilla SGD experience less catastrophic forgetting than the more modern algorithms such as Adam. We empirically compare four different existing metrics for quantifying catastrophic forgetting and (2) show that the degree to which the learning systems experience catastrophic forgetting is sufficiently sensitive to the metric used that a change from one principled metric to another is enough to change the conclusions of a study dramatically. Our results suggest that a much more rigorous experimental methodology is required when looking at catastrophic forgetting. Based on our results, we recommend inter-task forgetting in supervised learning must be measured with both retention and relearning metrics concurrently, and intra-task forgetting in reinforcement learning must-at the very least-be measured with pairwise interference.", "authors": ["Dylan R. Ashley", "Sina Ghiassian", "Richard S. Sutton"], "categories": ["cs.LG", "cs.AI", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2021-02-15", "url": "https://arxiv.org/abs/2102.07686", "pdf_url": "https://arxiv.org/pdf/2102.07686v4", "arxiv_id": "2102.07686", "doi": null, "citation_count": 9, "influential_citation_count": 1, "has_code": true, "code_url": "https://github.com/dylanashley/catastrophic-forgetting/tree/arxiv", "venue": null, "quality_score": 0.25}
{"id": "5987a4ca544db0665932633fb430a22f0fe5ff2bcba4b5c5a1b0e890a05bc104", "sources": ["arxiv", "semantic_scholar"], "title": "How do Quadratic Regularizers Prevent Catastrophic Forgetting: The Role of Interpolation", "abstract": "Catastrophic forgetting undermines the effectiveness of deep neural networks (DNNs) in scenarios such as continual learning and lifelong learning. While several methods have been proposed to tackle this problem, there is limited work explaining why these methods work well. This paper has the goal of better explaining a popularly used technique for avoiding catastrophic forgetting: quadratic regularization. We show that quadratic regularizers prevent forgetting of past tasks by interpolating current and previous values of model parameters at every training iteration. Over multiple training iterations, this interpolation operation reduces the learning rates of more important model parameters, thereby minimizing their movement. Our analysis also reveals two drawbacks of quadratic regularization: (a) dependence of parameter interpolation on training hyperparameters, which often leads to training instability and (b) assignment of lower importance to deeper layers, which are generally the place forgetting occurs in DNNs. Via a simple modification to the order of operations, we show these drawbacks can be easily avoided, resulting in 6.2\\% higher average accuracy at 4.5\\% lower average forgetting. We confirm the robustness of our results by training over 2000 models in different settings. Code available at \\url{https://github.com/EkdeepSLubana/QRforgetting}", "authors": ["Ekdeep Singh Lubana", "Puja Trivedi", "Danai Koutra", "Robert P. Dick"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2021-02-04", "url": "https://arxiv.org/abs/2102.02805", "pdf_url": "https://arxiv.org/pdf/2102.02805v5", "arxiv_id": "2102.02805", "doi": null, "citation_count": 19, "influential_citation_count": 2, "has_code": true, "code_url": "https://github.com/EkdeepSLubana/QRforgetting}", "venue": null, "quality_score": 0.3253}
{"id": "b656ea5efe362798d93d49f60d28ffacc551eb34462092244a52a1189e67a9b6", "sources": ["arxiv", "semantic_scholar"], "title": "Local Critic Training for Model-Parallel Learning of Deep Neural Networks", "abstract": "In this paper, we propose a novel model-parallel learning method, called local critic training, which trains neural networks using additional modules called local critic networks. The main network is divided into several layer groups and each layer group is updated through error gradients estimated by the corresponding local critic network. We show that the proposed approach successfully decouples the update process of the layer groups for both convolutional neural networks (CNNs) and recurrent neural networks (RNNs). In addition, we demonstrate that the proposed method is guaranteed to converge to a critical point. We also show that trained networks by the proposed method can be used for structural optimization. Experimental results show that our method achieves satisfactory performance, reduces training time greatly, and decreases memory consumption per machine. Code is available at https://github.com/hjdw2/Local-critic-training.", "authors": ["Hojung Lee", "Cho-Jui Hsieh", "Jong-Seok Lee"], "categories": ["cs.LG"], "fields_of_study": ["Medicine", "Computer Science"], "published_date": "2021-02-03", "url": "https://arxiv.org/abs/2102.01963", "pdf_url": "https://arxiv.org/pdf/2102.01963v1", "arxiv_id": "2102.01963", "doi": "10.1109/TNNLS.2021.3057380", "citation_count": 19, "influential_citation_count": 1, "has_code": true, "code_url": "https://github.com/hjdw2/Local-critic-training", "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.3253}
{"id": "4723e04e60371d1cc0c30c74da1624a30b33c45a484fbe096bccd83d6c19c8b8", "sources": ["arxiv", "semantic_scholar"], "title": "Online Continual Learning in Image Classification: An Empirical Survey", "abstract": "Online continual learning for image classification studies the problem of learning to classify images from an online stream of data and tasks, where tasks may include new classes (class incremental) or data nonstationarity (domain incremental). One of the key challenges of continual learning is to avoid catastrophic forgetting (CF), i.e., forgetting old tasks in the presence of more recent tasks. Over the past few years, many methods and tricks have been introduced to address this problem, but many have not been fairly and systematically compared under a variety of realistic and practical settings. To better understand the relative advantages of various approaches and the settings where they work best, this survey aims to (1) compare state-of-the-art methods such as MIR, iCARL, and GDumb and determine which works best at different experimental settings; (2) determine if the best class incremental methods are also competitive in domain incremental setting; (3) evaluate the performance of 7 simple but effective trick such as \"review\" trick and nearest class mean (NCM) classifier to assess their relative impact. Regarding (1), we observe iCaRL remains competitive when the memory buffer is small; GDumb outperforms many recently proposed methods in medium-size datasets and MIR performs the best in larger-scale datasets. For (2), we note that GDumb performs quite poorly while MIR -- already competitive for (1) -- is also strongly competitive in this very different but important setting. Overall, this allows us to conclude that MIR is overall a strong and versatile method across a wide variety of settings. For (3), we find that all 7 tricks are beneficial, and when augmented with the \"review\" trick and NCM classifier, MIR produces performance levels that bring online continual learning much closer to its ultimate goal of matching offline training.", "authors": ["Zheda Mai", "Ruiwen Li", "Jihwan Jeong", "David Quispe", "Hyunwoo Kim", "Scott Sanner"], "categories": ["cs.LG", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2021-01-25", "url": "https://arxiv.org/abs/2101.10423", "pdf_url": "https://arxiv.org/pdf/2101.10423v4", "arxiv_id": "2101.10423", "doi": "10.1016/j.neucom.2021.10.021", "citation_count": 520, "influential_citation_count": 45, "has_code": true, "code_url": "https://github.com/RaptorMai/online-continual-learning", "venue": "Neurocomputing", "quality_score": 0.8314}
{"id": "6ea65d21285bc406a3e29a1cca65a7ae43e7fc70cf712f95c91ab2ea779b41b8", "sources": ["arxiv", "semantic_scholar"], "title": "Learning Invariant Representation for Continual Learning", "abstract": "Continual learning aims to provide intelligent agents that are capable of learning continually a sequence of tasks, building on previously learned knowledge. A key challenge in this learning paradigm is catastrophically forgetting previously learned tasks when the agent faces a new one. Current rehearsal-based methods show their success in mitigating the catastrophic forgetting problem by replaying samples from previous tasks during learning a new one. However, these methods are infeasible when the data of previous tasks is not accessible. In this work, we propose a new pseudo-rehearsal-based method, named learning Invariant Representation for Continual Learning (IRCL), in which class-invariant representation is disentangled from a conditional generative model and jointly used with class-specific representation to learn the sequential tasks. Disentangling the shared invariant representation helps to learn continually a sequence of tasks, while being more robust to forgetting and having better knowledge transfer. We focus on class incremental learning where there is no knowledge about task identity during inference. We empirically evaluate our proposed method on two well-known benchmarks for continual learning: split MNIST and split Fashion MNIST. The experimental results show that our proposed method outperforms regularization-based methods by a big margin and is better than the state-of-the-art pseudo-rehearsal-based method. Finally, we analyze the role of the shared invariant representation in mitigating the forgetting problem especially when the number of replayed samples for each previous task is small.", "authors": ["Ghada Sokar", "Decebal Constantin Mocanu", "Mykola Pechenizkiy"], "categories": ["cs.LG", "cs.AI", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2021-01-15", "url": "https://arxiv.org/abs/2101.06162", "pdf_url": "https://arxiv.org/pdf/2101.06162v1", "arxiv_id": "2101.06162", "doi": null, "citation_count": 16, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3076}
{"id": "717163f298f717786ef8228ad3d619b838b50b7a1bf7a4d2fda3ebe778e9b7a8", "sources": ["arxiv", "semantic_scholar"], "title": "EEC: Learning to Encode and Regenerate Images for Continual Learning", "abstract": "The two main impediments to continual learning are catastrophic forgetting and memory limitations on the storage of data. To cope with these challenges, we propose a novel, cognitively-inspired approach which trains autoencoders with Neural Style Transfer to encode and store images. During training on a new task, reconstructed images from encoded episodes are replayed in order to avoid catastrophic forgetting. The loss function for the reconstructed images is weighted to reduce its effect during classifier training to cope with image degradation. When the system runs out of memory the encoded episodes are converted into centroids and covariance matrices, which are used to generate pseudo-images during classifier training, keeping classifier performance stable while using less memory. Our approach increases classification accuracy by 13-17% over state-of-the-art methods on benchmark datasets, while requiring 78% less storage space.", "authors": ["Ali Ayub", "Alan R. Wagner"], "categories": ["cs.CV", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2021-01-13", "url": "https://arxiv.org/abs/2101.04904", "pdf_url": "https://arxiv.org/pdf/2101.04904v4", "arxiv_id": "2101.04904", "doi": null, "citation_count": 63, "influential_citation_count": 3, "has_code": false, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.4515}
{"id": "57f415d05dd41f401cfc9b636090335f5a592bc921fb02114b61b6eb3071ea11", "sources": ["arxiv", "semantic_scholar"], "title": "Overcoming Catastrophic Forgetting in Graph Neural Networks", "abstract": "Catastrophic forgetting refers to the tendency that a neural network \"forgets\" the previous learned knowledge upon learning new tasks. Prior methods have been focused on overcoming this problem on convolutional neural networks (CNNs), where the input samples like images lie in a grid domain, but have largely overlooked graph neural networks (GNNs) that handle non-grid data. In this paper, we propose a novel scheme dedicated to overcoming catastrophic forgetting problem and hence strengthen continual learning in GNNs. At the heart of our approach is a generic module, termed as topology-aware weight preserving~(TWP), applicable to arbitrary form of GNNs in a plug-and-play fashion. Unlike the main stream of CNN-based continual learning methods that rely on solely slowing down the updates of parameters important to the downstream task, TWP explicitly explores the local structures of the input graph, and attempts to stabilize the parameters playing pivotal roles in the topological aggregation. We evaluate TWP on different GNN backbones over several datasets, and demonstrate that it yields performances superior to the state of the art. Code is publicly available at \\url{https://github.com/hhliu79/TWP}.", "authors": ["Huihui Liu", "Yiding Yang", "Xinchao Wang"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2020-12-10", "url": "https://arxiv.org/abs/2012.06002", "pdf_url": "https://arxiv.org/pdf/2012.06002v1", "arxiv_id": "2012.06002", "doi": "10.1609/aaai.v35i10.17049", "citation_count": 175, "influential_citation_count": 30, "has_code": true, "code_url": "https://github.com/hhliu79/TWP}", "venue": "AAAI Conference on Artificial Intelligence", "quality_score": 0.7457}
{"id": "6d53dba052c60671fcfb72204bd5e768c599811b6a845120e9c161c6f5f1921d", "sources": ["arxiv", "semantic_scholar"], "title": "Reset-Free Lifelong Learning with Skill-Space Planning", "abstract": "The objective of lifelong reinforcement learning (RL) is to optimize agents which can continuously adapt and interact in changing environments. However, current RL approaches fail drastically when environments are non-stationary and interactions are non-episodic. We propose Lifelong Skill Planning (LiSP), an algorithmic framework for non-episodic lifelong RL based on planning in an abstract space of higher-order skills. We learn the skills in an unsupervised manner using intrinsic rewards and plan over the learned skills using a learned dynamics model. Moreover, our framework permits skill discovery even from offline data, thereby reducing the need for excessive real-world interactions. We demonstrate empirically that LiSP successfully enables long-horizon planning and learns agents that can avoid catastrophic failures even in challenging non-stationary and non-episodic environments derived from gridworld and MuJoCo benchmarks.", "authors": ["Kevin Lu", "Aditya Grover", "Pieter Abbeel", "Igor Mordatch"], "categories": ["cs.LG", "cs.AI", "cs.RO"], "fields_of_study": ["Computer Science"], "published_date": "2020-12-07", "url": "https://arxiv.org/abs/2012.03548", "pdf_url": "https://arxiv.org/pdf/2012.03548v3", "arxiv_id": "2012.03548", "doi": null, "citation_count": 43, "influential_citation_count": 4, "has_code": false, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.4109}
{"id": "b2f8a6ec7747a44b0256cb545ae4899f98eb710f41b2efde1ba4e2a75841ca1c", "sources": ["arxiv", "semantic_scholar"], "title": "Model-Agnostic Learning to Meta-Learn", "abstract": "In this paper, we propose a learning algorithm that enables a model to quickly exploit commonalities among related tasks from an unseen task distribution, before quickly adapting to specific tasks from that same distribution. We investigate how learning with different task distributions can first improve adaptability by meta-finetuning on related tasks before improving goal task generalization with finetuning. Synthetic regression experiments validate the intuition that learning to meta-learn improves adaptability and consecutively generalization. Experiments on more complex image classification, continual regression, and reinforcement learning tasks demonstrate that learning to meta-learn generally improves task-specific adaptation. The methodology, setup, and hypotheses in this proposal were positively evaluated by peer review before conclusive experiments were carried out.", "authors": ["Arnout Devos", "Yatin Dandi"], "categories": ["cs.LG", "cs.AI", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-12-04", "url": "https://arxiv.org/abs/2012.02684", "pdf_url": "https://arxiv.org/pdf/2012.02684v2", "arxiv_id": "2012.02684", "doi": null, "citation_count": 3, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.1505}
{"id": "9b6117bb55d6931dc985f0b556a1d0a2ac5cb92a7f05fb383d17c9fa8a1caf6f", "sources": ["arxiv", "semantic_scholar"], "title": "Energy-Based Models for Continual Learning", "abstract": "We motivate Energy-Based Models (EBMs) as a promising model class for continual learning problems. Instead of tackling continual learning via the use of external memory, growing models, or regularization, EBMs change the underlying training objective to cause less interference with previously learned information. Our proposed version of EBMs for continual learning is simple, efficient, and outperforms baseline methods by a large margin on several benchmarks. Moreover, our proposed contrastive divergence-based training objective can be combined with other continual learning methods, resulting in substantial boosts in their performance. We further show that EBMs are adaptable to a more general continual learning setting where the data distribution changes without the notion of explicitly delineated tasks. These observations point towards EBMs as a useful building block for future continual learning methods.", "authors": ["Shuang Li", "Yilun Du", "Gido M. van de Ven", "Igor Mordatch"], "categories": ["cs.LG", "cs.AI", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-11-24", "url": "https://arxiv.org/abs/2011.12216", "pdf_url": "https://arxiv.org/pdf/2011.12216v3", "arxiv_id": "2011.12216", "doi": null, "citation_count": 49, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": "Proceedings of The 1st Conference on Lifelong Learning Agents, PMLR 199: 1-22, 2022", "quality_score": 0.4247}
{"id": "8c3948ded69b09f0cd102a8ea188cea6b7a19b0753b101c10db567f0593f60c3", "sources": ["arxiv", "semantic_scholar"], "title": "Generalized Continual Zero-Shot Learning", "abstract": "Recently, zero-shot learning (ZSL) emerged as an exciting topic and attracted a lot of attention. ZSL aims to classify unseen classes by transferring the knowledge from seen classes to unseen classes based on the class description. Despite showing promising performance, ZSL approaches assume that the training samples from all seen classes are available during the training, which is practically not feasible. To address this issue, we propose a more generalized and practical setup for ZSL, i.e., continual ZSL (CZSL), where classes arrive sequentially in the form of a task and it actively learns from the changing environment by leveraging the past experience. Further, to enhance the reliability, we develop CZSL for a single head continual learning setting where task identity is revealed during the training process but not during the testing. To avoid catastrophic forgetting and intransigence, we use knowledge distillation and storing and replay the few samples from previous tasks using a small episodic memory. We develop baselines and evaluate generalized CZSL on five ZSL benchmark datasets for two different settings of continual learning: with and without class incremental. Moreover, CZSL is developed for two types of variational autoencoders, which generates two types of features for classification: (i) generated features at output space and (ii) generated discriminative features at the latent space. The experimental results clearly indicate the single head CZSL is more generalizable and suitable for practical applications.", "authors": ["Chandan Gautam", "Sethupathy Parameswaran", "Ashish Mishra", "Suresh Sundaram"], "categories": ["cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2020-11-17", "url": "https://arxiv.org/abs/2011.08508", "pdf_url": "https://arxiv.org/pdf/2011.08508v3", "arxiv_id": "2011.08508", "doi": null, "citation_count": 12, "influential_citation_count": 3, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.301}
{"id": "40942244c31047aeb70589fbf77db5c250b67c7e1f09007b29862f1e83d31dfd", "sources": ["arxiv", "semantic_scholar"], "title": "Artificial Neural Variability for Deep Learning: On Overfitting, Noise Memorization, and Catastrophic Forgetting", "abstract": "Deep learning is often criticized by two serious issues which rarely exist in natural nervous systems: overfitting and catastrophic forgetting. It can even memorize randomly labelled data, which has little knowledge behind the instance-label pairs. When a deep network continually learns over time by accommodating new tasks, it usually quickly overwrites the knowledge learned from previous tasks. Referred to as the {\\it neural variability}, it is well-known in neuroscience that human brain reactions exhibit substantial variability even in response to the same stimulus. This mechanism balances accuracy and plasticity/flexibility in the motor learning of natural nervous systems. Thus it motivates us to design a similar mechanism named {\\it artificial neural variability} (ANV), which helps artificial neural networks learn some advantages from ``natural'' neural networks. We rigorously prove that ANV plays as an implicit regularizer of the mutual information between the training data and the learned model. This result theoretically guarantees ANV a strictly improved generalizability, robustness to label noise, and robustness to catastrophic forgetting. We then devise a {\\it neural variable risk minimization} (NVRM) framework and {\\it neural variable optimizers} to achieve ANV for conventional network architectures in practice. The empirical studies demonstrate that NVRM can effectively relieve overfitting, label noise memorization, and catastrophic forgetting at negligible costs. \\footnote{Code: \\url{https://github.com/zeke-xie/artificial-neural-variability-for-deep-learning}.", "authors": ["Zeke Xie", "Fengxiang He", "Shaopeng Fu", "Issei Sato", "Dacheng Tao", "Masashi Sugiyama"], "categories": ["cs.LG"], "fields_of_study": ["Medicine", "Computer Science"], "published_date": "2020-11-12", "url": "https://arxiv.org/abs/2011.06220", "pdf_url": "https://arxiv.org/pdf/2011.06220v3", "arxiv_id": "2011.06220", "doi": "10.1162/neco_a_01403", "citation_count": 70, "influential_citation_count": 1, "has_code": true, "code_url": "https://github.com/zeke-xie/artificial-neural-variability-for-deep-learning}", "venue": "Neural Computation", "quality_score": 0.4628}
{"id": "412ad1836b8b2f46854d24733bf726fd9703bc726e2d9e6ec828b76a930683cc", "sources": ["arxiv", "semantic_scholar"], "title": "Learning with Molecules beyond Graph Neural Networks", "abstract": "We demonstrate a deep learning framework which is inherently based in the highly expressive language of relational logic, enabling to, among other things, capture arbitrarily complex graph structures. We show how Graph Neural Networks and similar models can be easily covered in the framework by specifying the underlying propagation rules in the relational logic. The declarative nature of the used language then allows to easily modify and extend the propagation schemes into complex structures, such as the molecular rings which we choose for a short demonstration in this paper.", "authors": ["Gustav Sourek", "Filip Zelezny", "Ondrej Kuzelka"], "categories": ["cs.LG", "cs.AI", "cs.LO", "cs.NE"], "fields_of_study": ["Computer Science"], "published_date": "2020-11-06", "url": "https://arxiv.org/abs/2011.03488", "pdf_url": "https://arxiv.org/pdf/2011.03488v1", "arxiv_id": "2011.03488", "doi": null, "citation_count": 3, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1505}
{"id": "e48464867f1ca1926e27025dc203839ae44cbffda76dbb5993b4976eb1970063", "sources": ["arxiv", "semantic_scholar"], "title": "Learning Invariances in Neural Networks", "abstract": "Invariances to translations have imbued convolutional neural networks with powerful generalization properties. However, we often do not know a priori what invariances are present in the data, or to what extent a model should be invariant to a given symmetry group. We show how to \\emph{learn} invariances and equivariances by parameterizing a distribution over augmentations and optimizing the training loss simultaneously with respect to the network parameters and augmentation parameters. With this simple procedure we can recover the correct set and extent of invariances on image classification, regression, segmentation, and molecular property prediction from a large space of augmentations, on training data alone.", "authors": ["Gregory Benton", "Marc Finzi", "Pavel Izmailov", "Andrew Gordon Wilson"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-10-22", "url": "https://arxiv.org/abs/2010.11882", "pdf_url": "https://arxiv.org/pdf/2010.11882v2", "arxiv_id": "2010.11882", "doi": null, "citation_count": 79, "influential_citation_count": 15, "has_code": true, "code_url": "https://github.com/g-benton/learning-invariances", "venue": "Neural Information Processing Systems", "quality_score": 0.6021}
{"id": "636fdb24579e511324e0e173761210eff991b3748efae003124463d2a2bb159c", "sources": ["arxiv", "semantic_scholar"], "title": "Deep Reinforcement Learning with Population-Coded Spiking Neural Network for Continuous Control", "abstract": "The energy-efficient control of mobile robots is crucial as the complexity of their real-world applications increasingly involves high-dimensional observation and action spaces, which cannot be offset by limited on-board resources. An emerging non-Von Neumann model of intelligence, where spiking neural networks (SNNs) are run on neuromorphic processors, is regarded as an energy-efficient and robust alternative to the state-of-the-art real-time robotic controllers for low dimensional control tasks. The challenge now for this new computing paradigm is to scale so that it can keep up with real-world tasks. To do so, SNNs need to overcome the inherent limitations of their training, namely the limited ability of their spiking neurons to represent information and the lack of effective learning algorithms. Here, we propose a population-coded spiking actor network (PopSAN) trained in conjunction with a deep critic network using deep reinforcement learning (DRL). The population coding scheme dramatically increased the representation capacity of the network and the hybrid learning combined the training advantages of deep networks with the energy-efficient inference of spiking networks. To show the general applicability of our approach, we integrated it with a spectrum of both on-policy and off-policy DRL algorithms. We deployed the trained PopSAN on Intel's Loihi neuromorphic chip and benchmarked our method against the mainstream DRL algorithms for continuous control. To allow for a fair comparison among all methods, we validated them on OpenAI gym tasks. Our Loihi-run PopSAN consumed 140 times less energy per inference when compared against the deep actor network on Jetson TX2, and had the same level of performance. Our results support the efficiency of neuromorphic controllers and suggest our hybrid RL as an alternative to deep learning, when both energy-efficiency and robustness are important.", "authors": ["Guangzhi Tang", "Neelesh Kumar", "Raymond Yoo", "Konstantinos P. Michmizos"], "categories": ["cs.NE", "cs.LG", "cs.RO"], "fields_of_study": ["Computer Science"], "published_date": "2020-10-19", "url": "https://arxiv.org/abs/2010.09635", "pdf_url": "https://arxiv.org/pdf/2010.09635v1", "arxiv_id": "2010.09635", "doi": null, "citation_count": 120, "influential_citation_count": 18, "has_code": false, "code_url": null, "venue": "Conference on Robot Learning", "quality_score": 0.6394}
{"id": "459d8b194fb93e0a2173c0c8aa8fc08f5117d80a591ba7d2b673cdc1da9fc404", "sources": ["arxiv", "semantic_scholar"], "title": "A Theoretical Analysis of Catastrophic Forgetting through the NTK Overlap Matrix", "abstract": "Continual learning (CL) is a setting in which an agent has to learn from an incoming stream of data during its entire lifetime. Although major advances have been made in the field, one recurring problem which remains unsolved is that of Catastrophic Forgetting (CF). While the issue has been extensively studied empirically, little attention has been paid from a theoretical angle. In this paper, we show that the impact of CF increases as two tasks increasingly align. We introduce a measure of task similarity called the NTK overlap matrix which is at the core of CF. We analyze common projected gradient algorithms and demonstrate how they mitigate forgetting. Then, we propose a variant of Orthogonal Gradient Descent (OGD) which leverages structure of the data through Principal Component Analysis (PCA). Experiments support our theoretical findings and show how our method can help reduce CF on classical CL datasets.", "authors": ["Thang Doan", "Mehdi Bennani", "Bogdan Mazoure", "Guillaume Rabusseau", "Pierre Alquier"], "categories": ["cs.LG", "cs.AI", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-10-07", "url": "https://arxiv.org/abs/2010.04003", "pdf_url": "https://arxiv.org/pdf/2010.04003v2", "arxiv_id": "2010.04003", "doi": null, "citation_count": 115, "influential_citation_count": 12, "has_code": false, "code_url": null, "venue": "International Conference on Artificial Intelligence and Statistics", "quality_score": 0.557}
{"id": "42cc21765d01f60d1568575bbf0059daf6d5a03e454fb67466b1003f7fcc62d9", "sources": ["arxiv", "semantic_scholar"], "title": "Task Agnostic Continual Learning Using Online Variational Bayes with Fixed-Point Updates", "abstract": "Background: Catastrophic forgetting is the notorious vulnerability of neural networks to the changes in the data distribution during learning. This phenomenon has long been considered a major obstacle for using learning agents in realistic continual learning settings. A large body of continual learning research assumes that task boundaries are known during training. However, only a few works consider scenarios in which task boundaries are unknown or not well defined -- task agnostic scenarios. The optimal Bayesian solution for this requires an intractable online Bayes update to the weights posterior. Contributions: We aim to approximate the online Bayes update as accurately as possible. To do so, we derive novel fixed-point equations for the online variational Bayes optimization problem, for multivariate Gaussian parametric distributions. By iterating the posterior through these fixed-point equations, we obtain an algorithm (FOO-VB) for continual learning which can handle non-stationary data distribution using a fixed architecture and without using external memory (i.e. without access to previous data). We demonstrate that our method (FOO-VB) outperforms existing methods in task agnostic scenarios. FOO-VB Pytorch implementation will be available online.", "authors": ["Chen Zeno", "Itay Golan", "Elad Hoffer", "Daniel Soudry"], "categories": ["stat.ML", "cs.LG"], "fields_of_study": ["Computer Science", "Medicine", "Mathematics"], "published_date": "2020-10-01", "url": "https://arxiv.org/abs/2010.00373", "pdf_url": "https://arxiv.org/pdf/2010.00373v2", "arxiv_id": "2010.00373", "doi": "10.1162/neco_a_01430", "citation_count": 52, "influential_citation_count": 4, "has_code": false, "code_url": null, "venue": "Neural Computation", "quality_score": 0.4311}
{"id": "ad03857bd8cdcd237fa2b2f91f01dd6c91d04a73b58846d2be46ed6a394f9134", "sources": ["arxiv", "semantic_scholar"], "title": "Beneficial Perturbation Network for designing general adaptive artificial intelligence systems", "abstract": "The human brain is the gold standard of adaptive learning. It not only can learn and benefit from experience, but also can adapt to new situations. In contrast, deep neural networks only learn one sophisticated but fixed mapping from inputs to outputs. This limits their applicability to more dynamic situations, where input to output mapping may change with different contexts. A salient example is continual learning - learning new independent tasks sequentially without forgetting previous tasks. Continual learning of multiple tasks in artificial neural networks using gradient descent leads to catastrophic forgetting, whereby a previously learned mapping of an old task is erased when learning new mappings for new tasks. Here, we propose a new biologically plausible type of deep neural network with extra, out-of-network, task-dependent biasing units to accommodate these dynamic situations. This allows, for the first time, a single network to learn potentially unlimited parallel input to output mappings, and to switch on the fly between them at runtime. Biasing units are programmed by leveraging beneficial perturbations (opposite to well-known adversarial perturbations) for each task. Beneficial perturbations for a given task bias the network toward that task, essentially switching the network into a different mode to process that task. This largely eliminates catastrophic interference between tasks. Our approach is memory-efficient and parameter-efficient, can accommodate many tasks, and achieves state-of-the-art performance across different tasks and domains.", "authors": ["Shixian Wen", "Amanda Rios", "Yunhao Ge", "Laurent Itti"], "categories": ["cs.CV", "cs.AI"], "fields_of_study": ["Computer Science", "Medicine"], "published_date": "2020-09-27", "url": "https://arxiv.org/abs/2009.13954", "pdf_url": "https://arxiv.org/pdf/2009.13954v2", "arxiv_id": "2009.13954", "doi": "10.1109/TNNLS.2021.3054423", "citation_count": 18, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.3197}
{"id": "cdb32c2bd659d511059508210f9f078b2a3f99780428bf1c7219dc52761d6ddc", "sources": ["arxiv", "semantic_scholar"], "title": "Theoretical Analysis of the Advantage of Deepening Neural Networks", "abstract": "We propose two new criteria to understand the advantage of deepening neural networks. It is important to know the expressivity of functions computable by deep neural networks in order to understand the advantage of deepening neural networks. Unless deep neural networks have enough expressivity, they cannot have good performance even though learning is successful. In this situation, the proposed criteria contribute to understanding the advantage of deepening neural networks since they can evaluate the expressivity independently from the efficiency of learning. The first criterion shows the approximation accuracy of deep neural networks to the target function. This criterion has the background that the goal of deep learning is approximating the target function by deep neural networks. The second criterion shows the property of linear regions of functions computable by deep neural networks. This criterion has the background that deep neural networks whose activation functions are piecewise linear are also piecewise linear. Furthermore, by the two criteria, we show that to increase layers is more effective than to increase units at each layer on improving the expressivity of deep neural networks.", "authors": ["Yasushi Esaki", "Yuta Nakahara", "Toshiyasu Matsushima"], "categories": ["cs.LG", "cs.NE", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-09-24", "url": "https://arxiv.org/abs/2009.11479", "pdf_url": "https://arxiv.org/pdf/2009.11479v1", "arxiv_id": "2009.11479", "doi": "10.1109/ICMLA51294.2020.00081", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning and Applications", "quality_score": 0.0753}
{"id": "a9f48fd186c794d2131c1e9e1463d6c7b70906f41249d0c3aff05fe6132bada1", "sources": ["arxiv", "semantic_scholar"], "title": "Online Learning With Adaptive Rebalancing in Nonstationary Environments", "abstract": "An enormous and ever-growing volume of data is nowadays becoming available in a sequential fashion in various real-world applications. Learning in nonstationary environments constitutes a major challenge, and this problem becomes orders of magnitude more complex in the presence of class imbalance. We provide new insights into learning from nonstationary and imbalanced data in online learning, a largely unexplored area. We propose the novel Adaptive REBAlancing (AREBA) algorithm that selectively includes in the training set a subset of the majority and minority examples that appeared so far, while at its heart lies an adaptive mechanism to continually maintain the class balance between the selected examples. We compare AREBA with strong baselines and other state-of-the-art algorithms and perform extensive experimental work in scenarios with various class imbalance rates and different concept drift types on both synthetic and real-world data. AREBA significantly outperforms the rest with respect to both learning speed and learning quality. Our code is made publicly available to the scientific community.", "authors": ["Kleanthis Malialis", "Christos G. Panayiotou", "Marios M. Polycarpou"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Medicine", "Computer Science", "Mathematics"], "published_date": "2020-09-24", "url": "https://arxiv.org/abs/2009.11942", "pdf_url": "https://arxiv.org/pdf/2009.11942v1", "arxiv_id": "2009.11942", "doi": "10.1109/TNNLS.2020.3017863", "citation_count": 47, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.4203}
{"id": "9803d1f7adcbb0946490a26bd1cf44f523fc48dc9a21f8001312df209f884b65", "sources": ["arxiv", "semantic_scholar"], "title": "Anomalous diffusion dynamics of learning in deep neural networks", "abstract": "Learning in deep neural networks (DNNs) is implemented through minimizing a highly non-convex loss function, typically by a stochastic gradient descent (SGD) method. This learning process can effectively find good wide minima without being trapped in poor local ones. We present a novel account of how such effective deep learning emerges through the interactions of the SGD and the geometrical structure of the loss landscape. Rather than being a normal diffusion process (i.e. Brownian motion) as often assumed, we find that the SGD exhibits rich, complex dynamics when navigating through the loss landscape; initially, the SGD exhibits anomalous superdiffusion, which attenuates gradually and changes to subdiffusion at long times when the solution is reached. Such learning dynamics happen ubiquitously in different DNNs such as ResNet and VGG-like networks and are insensitive to batch size and learning rate. The anomalous superdiffusion process during the initial learning phase indicates that the motion of SGD along the loss landscape possesses intermittent, big jumps; this non-equilibrium property enables the SGD to escape from sharp local minima. By adapting the methods developed for studying energy landscapes in complex physical systems, we find that such superdiffusive learning dynamics are due to the interactions of the SGD and the fractal-like structure of the loss landscape. We further develop a simple model to demonstrate the mechanistic role of the fractal loss landscape in enabling the SGD to effectively find global minima. Our results thus reveal the effectiveness of deep learning from a novel perspective and have implications for designing efficient deep neural networks.", "authors": ["Guozhang Chen", "Cheng Kevin Qu", "Pulin Gong"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Medicine", "Mathematics"], "published_date": "2020-09-22", "url": "https://arxiv.org/abs/2009.10588", "pdf_url": "https://arxiv.org/pdf/2009.10588v2", "arxiv_id": "2009.10588", "doi": "10.1016/j.neunet.2022.01.019", "citation_count": 27, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "Neural Networks", "quality_score": 0.3618}
{"id": "ea31c7998913fbd1f0522ee1b75267f1d28588e0f9ed334293ef3bcc167e69d1", "sources": ["arxiv", "semantic_scholar"], "title": "Few-Shot Unsupervised Continual Learning through Meta-Examples", "abstract": "In real-world applications, data do not reflect the ones commonly used for neural networks training, since they are usually few, unlabeled and can be available as a stream. Hence many existing deep learning solutions suffer from a limited range of applications, in particular in the case of online streaming data that evolve over time. To narrow this gap, in this work we introduce a novel and complex setting involving unsupervised meta-continual learning with unbalanced tasks. These tasks are built through a clustering procedure applied to a fitted embedding space. We exploit a meta-learning scheme that simultaneously alleviates catastrophic forgetting and favors the generalization to new tasks. Moreover, to encourage feature reuse during the meta-optimization, we exploit a single inner loop taking advantage of an aggregated representation achieved through the use of a self-attention mechanism. Experimental results on few-shot learning benchmarks show competitive performance even compared to the supervised case. Additionally, we empirically observe that in an unsupervised scenario, the small tasks and the variability in the clusters pooling play a crucial role in the generalization capability of the network. Further, on complex datasets, the exploitation of more clusters than the true number of classes leads to higher results, even compared to the ones obtained with full supervision, suggesting that a predefined partitioning into classes can miss relevant structural information.", "authors": ["Alessia Bertugli", "Stefano Vincenzi", "Simone Calderara", "Andrea Passerini"], "categories": ["cs.LG", "cs.CV", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-09-17", "url": "https://arxiv.org/abs/2009.08107", "pdf_url": "https://arxiv.org/pdf/2009.08107v3", "arxiv_id": "2009.08107", "doi": null, "citation_count": 8, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2386}
{"id": "ae29b1cad21cae8a1cf2b3743fac60a28416e76ff30ddad3bc76a77689db93f2", "sources": ["arxiv", "semantic_scholar"], "title": "Routing Networks with Co-training for Continual Learning", "abstract": "The core challenge with continual learning is catastrophic forgetting, the phenomenon that when neural networks are trained on a sequence of tasks they rapidly forget previously learned tasks. It has been observed that catastrophic forgetting is most severe when tasks are dissimilar to each other. We propose the use of sparse routing networks for continual learning. For each input, these network architectures activate a different path through a network of experts. Routing networks have been shown to learn to route similar tasks to overlapping sets of experts and dissimilar tasks to disjoint sets of experts. In the continual learning context this behaviour is desirable as it minimizes interference between dissimilar tasks while allowing positive transfer between related tasks. In practice, we find it is necessary to develop a new training method for routing networks, which we call co-training which avoids poorly initialized experts when new tasks are presented. When combined with a small episodic memory replay buffer, sparse routing networks with co-training outperform densely connected networks on the MNIST-Permutations and MNIST-Rotations benchmarks.", "authors": ["Mark Collier", "Efi Kokiopoulou", "Andrea Gesmundo", "Jesse Berent"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-09-09", "url": "https://arxiv.org/abs/2009.04381", "pdf_url": "https://arxiv.org/pdf/2009.04381v1", "arxiv_id": "2009.04381", "doi": null, "citation_count": 15, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.301}
{"id": "e59fc03c9de24712a1263328ab7eb64ab8ba55f95e790d35881a31254b46a3ca", "sources": ["arxiv", "semantic_scholar"], "title": "A Wholistic View of Continual Learning with Deep Neural Networks: Forgotten Lessons and the Bridge to Active and Open World Learning", "abstract": "Current deep learning methods are regarded as favorable if they empirically perform well on dedicated test sets. This mentality is seamlessly reflected in the resurfacing area of continual learning, where consecutively arriving data is investigated. The core challenge is framed as protecting previously acquired representations from being catastrophically forgotten. However, comparison of individual methods is nevertheless performed in isolation from the real world by monitoring accumulated benchmark test set performance. The closed world assumption remains predominant, i.e. models are evaluated on data that is guaranteed to originate from the same distribution as used for training. This poses a massive challenge as neural networks are well known to provide overconfident false predictions on unknown and corrupted instances. In this work we critically survey the literature and argue that notable lessons from open set recognition, identifying unknown examples outside of the observed set, and the adjacent field of active learning, querying data to maximize the expected performance gain, are frequently overlooked in the deep learning era. Hence, we propose a consolidated view to bridge continual learning, active learning and open set recognition in deep neural networks. Finally, the established synergies are supported empirically, showing joint improvement in alleviating catastrophic forgetting, querying data, selecting task orders, while exhibiting robust open world application.", "authors": ["Martin Mundt", "Yongwon Hong", "Iuliia Pliushch", "Visvanathan Ramesh"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Medicine", "Mathematics"], "published_date": "2020-09-03", "url": "https://arxiv.org/abs/2009.01797", "pdf_url": "https://arxiv.org/pdf/2009.01797v3", "arxiv_id": "2009.01797", "doi": "10.1016/j.neunet.2023.01.014", "citation_count": 182, "influential_citation_count": 8, "has_code": false, "code_url": null, "venue": "Neural Networks", "quality_score": 0.5656}
{"id": "36750662c20d270de6e87816af0e39ddfdc6a93d460e879b3a3a79de902d52a4", "sources": ["arxiv", "semantic_scholar"], "title": "Lifelong Graph Learning", "abstract": "Graph neural networks (GNN) are powerful models for many graph-structured tasks. Existing models often assume that the complete structure of the graph is available during training. In practice, however, graph-structured data is usually formed in a streaming fashion so that learning a graph continuously is often necessary. In this paper, we bridge GNN and lifelong learning by converting a continual graph learning problem to a regular graph learning problem so GNN can inherit the lifelong learning techniques developed for convolutional neural networks (CNN). We propose a new topology, the feature graph, which takes features as new nodes and turns nodes into independent graphs. This successfully converts the original problem of node classification to graph classification. In the experiments, we demonstrate the efficiency and effectiveness of feature graph networks (FGN) by continuously learning a sequence of classical graph datasets. We also show that FGN achieves superior performance in two applications, i.e., lifelong human action recognition with wearable devices and feature matching. To the best of our knowledge, FGN is the first method to bridge graph learning and lifelong learning via a novel graph topology. Source code is available at https://github.com/wang-chen/LGL", "authors": ["Chen Wang", "Yuheng Qiu", "Dasong Gao", "Sebastian Scherer"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-09-01", "url": "https://arxiv.org/abs/2009.00647", "pdf_url": "https://arxiv.org/pdf/2009.00647v4", "arxiv_id": "2009.00647", "doi": "10.1109/CVPR52688.2022.01335", "citation_count": 61, "influential_citation_count": 8, "has_code": true, "code_url": "https://github.com/wang-chen/LGL", "venue": "Computer Vision and Pattern Recognition", "quality_score": 0.4771}
{"id": "eb42acf3efba04d6b5caa3bd772cbe08cb0a788d7aa26a9c31acfd70b3d1d570", "sources": ["arxiv", "semantic_scholar"], "title": "Amortized learning of neural causal representations", "abstract": "Causal models can compactly and efficiently encode the data-generating process under all interventions and hence may generalize better under changes in distribution. These models are often represented as Bayesian networks and learning them scales poorly with the number of variables. Moreover, these approaches cannot leverage previously learned knowledge to help with learning new causal models. In order to tackle these challenges, we represent a novel algorithm called \\textit{causal relational networks} (CRN) for learning causal models using neural networks. The CRN represent causal models using continuous representations and hence could scale much better with the number of variables. These models also take in previously learned information to facilitate learning of new causal models. Finally, we propose a decoding-based metric to evaluate causal models with continuous representations. We test our method on synthetic data achieving high accuracy and quick adaptation to previously unseen causal models.", "authors": ["Nan Rosemary Ke", "Jane. X. Wang", "Jovana Mitrovic", "Martin Szummer", "Danilo J. Rezende"], "categories": ["stat.ML", "cs.LG"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-08-21", "url": "https://arxiv.org/abs/2008.09301", "pdf_url": "https://arxiv.org/pdf/2008.09301v1", "arxiv_id": "2008.09301", "doi": null, "citation_count": 22, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3404}
{"id": "ec5ba80fac68c7b04c55f4868179f4f35353e3bd2589aabfa58fffe4dab007b8", "sources": ["arxiv", "semantic_scholar"], "title": "Interactive Imitation Learning in State-Space", "abstract": "Imitation Learning techniques enable programming the behavior of agents through demonstrations rather than manual engineering. However, they are limited by the quality of available demonstration data. Interactive Imitation Learning techniques can improve the efficacy of learning since they involve teachers providing feedback while the agent executes its task. In this work, we propose a novel Interactive Learning technique that uses human feedback in state-space to train and improve agent behavior (as opposed to alternative methods that use feedback in action-space). Our method titled Teaching Imitative Policies in State-space~(TIPS) enables providing guidance to the agent in terms of `changing its state' which is often more intuitive for a human demonstrator. Through continuous improvement via corrective feedback, agents trained by non-expert demonstrators using TIPS outperformed the demonstrator and conventional Imitation Learning agents.", "authors": ["Snehal Jauhri", "Carlos Celemin", "Jens Kober"], "categories": ["cs.RO", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2020-08-02", "url": "https://arxiv.org/abs/2008.00524", "pdf_url": "https://arxiv.org/pdf/2008.00524v2", "arxiv_id": "2008.00524", "doi": null, "citation_count": 16, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "Conference on Robot Learning", "quality_score": 0.3076}
{"id": "41ddf9b65d0ef986e2f7fe8a25b011f27925a03a84001063556134b6398a9572", "sources": ["arxiv", "semantic_scholar"], "title": "Lifelong Incremental Reinforcement Learning with Online Bayesian Inference", "abstract": "A central capability of a long-lived reinforcement learning (RL) agent is to incrementally adapt its behavior as its environment changes, and to incrementally build upon previous experiences to facilitate future learning in real-world scenarios. In this paper, we propose LifeLong Incremental Reinforcement Learning (LLIRL), a new incremental algorithm for efficient lifelong adaptation to dynamic environments. We develop and maintain a library that contains an infinite mixture of parameterized environment models, which is equivalent to clustering environment parameters in a latent space. The prior distribution over the mixture is formulated as a Chinese restaurant process (CRP), which incrementally instantiates new environment models without any external information to signal environmental changes in advance. During lifelong learning, we employ the expectation maximization (EM) algorithm with online Bayesian inference to update the mixture in a fully incremental manner. In EM, the E-step involves estimating the posterior expectation of environment-to-cluster assignments, while the M-step updates the environment parameters for future learning. This method allows for all environment models to be adapted as necessary, with new models instantiated for environmental changes and old models retrieved when previously seen environments are encountered again. Experiments demonstrate that LLIRL outperforms relevant existing methods, and enables effective incremental adaptation to various dynamic environments for lifelong learning.", "authors": ["Zhi Wang", "Chunlin Chen", "Daoyi Dong"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science", "Medicine"], "published_date": "2020-07-28", "url": "https://arxiv.org/abs/2007.14196", "pdf_url": "https://arxiv.org/pdf/2007.14196v2", "arxiv_id": "2007.14196", "doi": "10.1109/TNNLS.2021.3055499", "citation_count": 67, "influential_citation_count": 6, "has_code": false, "code_url": null, "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.4581}
{"id": "ff7965e5814e9639a868918e7a374c0cea96bb398b2f3cc9d480bb880e9066cc", "sources": ["arxiv", "semantic_scholar"], "title": "Tighter risk certificates for neural networks", "abstract": "This paper presents an empirical study regarding training probabilistic neural networks using training objectives derived from PAC-Bayes bounds. In the context of probabilistic neural networks, the output of training is a probability distribution over network weights. We present two training objectives, used here for the first time in connection with training neural networks. These two training objectives are derived from tight PAC-Bayes bounds. We also re-implement a previously used training objective based on a classical PAC-Bayes bound, to compare the properties of the predictors learned using the different training objectives. We compute risk certificates for the learnt predictors, based on part of the data used to learn the predictors. We further experiment with different types of priors on the weights (both data-free and data-dependent priors) and neural network architectures. Our experiments on MNIST and CIFAR-10 show that our training methods produce competitive test set errors and non-vacuous risk bounds with much tighter values than previous results in the literature, showing promise not only to guide the learning algorithm through bounding the risk but also for model selection. These observations suggest that the methods studied here might be good candidates for self-certified learning, in the sense of using the whole data set for learning a predictor and certifying its risk on any unseen data (from the same distribution as the training data) potentially without the need for holding out test data.", "authors": ["María Pérez-Ortiz", "Omar Rivasplata", "John Shawe-Taylor", "Csaba Szepesvári"], "categories": ["cs.LG", "cs.CV", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-07-25", "url": "https://arxiv.org/abs/2007.12911", "pdf_url": "https://arxiv.org/pdf/2007.12911v3", "arxiv_id": "2007.12911", "doi": null, "citation_count": 130, "influential_citation_count": 26, "has_code": false, "code_url": null, "venue": "Journal of machine learning research", "quality_score": 0.7157}
{"id": "c2ab9f52deb23a080b265ae7c85af31d6eae769f6647c6bfd9e6df0c3adf44f4", "sources": ["arxiv", "semantic_scholar"], "title": "Mind Your Manners! A Dataset and A Continual Learning Approach for Assessing Social Appropriateness of Robot Actions", "abstract": "To date, endowing robots with an ability to assess social appropriateness of their actions has not been possible. This has been mainly due to (i) the lack of relevant and labelled data, and (ii) the lack of formulations of this as a lifelong learning problem. In this paper, we address these two issues. We first introduce the Socially Appropriate Domestic Robot Actions dataset (MANNERS-DB), which contains appropriateness labels of robot actions annotated by humans. To be able to control but vary the configurations of the scenes and the social settings, MANNERS-DB has been created utilising a simulation environment by uniformly sampling relevant contextual attributes. Secondly, we train and evaluate a baseline Bayesian Neural Network (BNN) that estimates social appropriateness of actions in the MANNERS-DB. Finally, we formulate learning social appropriateness of actions as a continual learning problem using the uncertainty of the BNN parameters. The experimental results show that the social appropriateness of robot actions can be predicted with a satisfactory level of precision. Our work takes robots one step closer to a human-like understanding of (social) appropriateness of actions, with respect to the social context they operate in. To facilitate reproducibility and further progress in this area, the MANNERS-DB, the trained models and the relevant code will be made publicly available.", "authors": ["Jonas Tjomsland", "Sinan Kalkan", "Hatice Gunes"], "categories": ["cs.RO", "cs.HC"], "fields_of_study": ["Medicine", "Computer Science"], "published_date": "2020-07-24", "url": "https://arxiv.org/abs/2007.12506", "pdf_url": "https://arxiv.org/pdf/2007.12506v1", "arxiv_id": "2007.12506", "doi": "10.3389/frobt.2022.669420", "citation_count": 25, "influential_citation_count": 4, "has_code": false, "code_url": null, "venue": "Frontiers in Robotics and AI", "quality_score": 0.3537}
{"id": "04e7b0c4a2fab1ead43d2e5787b965211783b7b790fe0ab517e809ed3a124487", "sources": ["arxiv", "semantic_scholar"], "title": "Lifelong Learning of Compositional Structures", "abstract": "A hallmark of human intelligence is the ability to construct self-contained chunks of knowledge and adequately reuse them in novel combinations for solving different yet structurally related problems. Learning such compositional structures has been a significant challenge for artificial systems, due to the combinatorial nature of the underlying search problem. To date, research into compositional learning has largely proceeded separately from work on lifelong or continual learning. We integrate these two lines of work to present a general-purpose framework for lifelong learning of compositional structures that can be used for solving a stream of related tasks. Our framework separates the learning process into two broad stages: learning how to best combine existing components in order to assimilate a novel problem, and learning how to adapt the set of existing components to accommodate the new problem. This separation explicitly handles the trade-off between the stability required to remember how to solve earlier tasks and the flexibility required to solve new tasks, as we show empirically in an extensive evaluation.", "authors": ["Jorge A. Mendez", "Eric Eaton"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-07-15", "url": "https://arxiv.org/abs/2007.07732", "pdf_url": "https://arxiv.org/pdf/2007.07732v2", "arxiv_id": "2007.07732", "doi": null, "citation_count": 51, "influential_citation_count": 2, "has_code": true, "code_url": "https://github.com/Lifelong-ML/Mendez2020Compositional.git", "venue": "International Conference on Learning Representations", "quality_score": 0.429}
{"id": "39e692658d597d3565538c7d2301561bde8474d6c5a72bf7008fab592986c90f", "sources": ["arxiv", "semantic_scholar"], "title": "Lifelong Learning using Eigentasks: Task Separation, Skill Acquisition, and Selective Transfer", "abstract": "We introduce the eigentask framework for lifelong learning. An eigentask is a pairing of a skill that solves a set of related tasks, paired with a generative model that can sample from the skill's input space. The framework extends generative replay approaches, which have mainly been used to avoid catastrophic forgetting, to also address other lifelong learning goals such as forward knowledge transfer. We propose a wake-sleep cycle of alternating task learning and knowledge consolidation for learning in our framework, and instantiate it for lifelong supervised learning and lifelong RL. We achieve improved performance over the state-of-the-art in supervised continual learning, and show evidence of forward knowledge transfer in a lifelong RL application in the game Starcraft2.", "authors": ["Aswin Raghavan", "Jesse Hostetler", "Indranil Sur", "Abrar Rahman", "Ajay Divakaran"], "categories": ["cs.LG", "cs.AI", "cs.CV", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-07-14", "url": "https://arxiv.org/abs/2007.06918", "pdf_url": "https://arxiv.org/pdf/2007.06918v1", "arxiv_id": "2007.06918", "doi": null, "citation_count": 7, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2258}
{"id": "655e95e83f0e3eeba71e544206d9edb5cdfd20ac841fe8e7d86a5c867b81e082", "sources": ["arxiv", "semantic_scholar"], "title": "Lifelong Policy Gradient Learning of Factored Policies for Faster Training Without Forgetting", "abstract": "Policy gradient methods have shown success in learning control policies for high-dimensional dynamical systems. Their biggest downside is the amount of exploration they require before yielding high-performing policies. In a lifelong learning setting, in which an agent is faced with multiple consecutive tasks over its lifetime, reusing information from previously seen tasks can substantially accelerate the learning of new tasks. We provide a novel method for lifelong policy gradient learning that trains lifelong function approximators directly via policy gradients, allowing the agent to benefit from accumulated knowledge throughout the entire training process. We show empirically that our algorithm learns faster and converges to better policies than single-task and lifelong learning baselines, and completely avoids catastrophic forgetting on a variety of challenging domains.", "authors": ["Jorge A. Mendez", "Boyu Wang", "Eric Eaton"], "categories": ["cs.LG", "cs.AI", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-07-14", "url": "https://arxiv.org/abs/2007.07011", "pdf_url": "https://arxiv.org/pdf/2007.07011v2", "arxiv_id": "2007.07011", "doi": null, "citation_count": 42, "influential_citation_count": 5, "has_code": false, "code_url": null, "venue": "Neural Information Processing Systems", "quality_score": 0.4084}
{"id": "131df23a12abf1b3b8cadaa44b4beb9dfa241f4870eba7d4f4becd99904725f2", "sources": ["arxiv", "semantic_scholar"], "title": "VAFL: a Method of Vertical Asynchronous Federated Learning", "abstract": "Horizontal Federated learning (FL) handles multi-client data that share the same set of features, and vertical FL trains a better predictor that combine all the features from different clients. This paper targets solving vertical FL in an asynchronous fashion, and develops a simple FL method. The new method allows each client to run stochastic gradient algorithms without coordination with other clients, so it is suitable for intermittent connectivity of clients. This method further uses a new technique of perturbed local embedding to ensure data privacy and improve communication efficiency. Theoretically, we present the convergence rate and privacy level of our method for strongly convex, nonconvex and even nonsmooth objectives separately. Empirically, we apply our method to FL on various image and healthcare datasets. The results compare favorably to centralized and synchronous FL methods.", "authors": ["Tianyi Chen", "Xiao Jin", "Yuejiao Sun", "Wotao Yin"], "categories": ["cs.LG", "cs.DC", "math.OC", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-07-12", "url": "https://arxiv.org/abs/2007.06081", "pdf_url": "https://arxiv.org/pdf/2007.06081v1", "arxiv_id": "2007.06081", "doi": null, "citation_count": 196, "influential_citation_count": 30, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.7457}
{"id": "d379033f3c6a23d968bd88b0577c0485bb5576df6d40cc6838bc1ff79f7c915e", "sources": ["arxiv", "semantic_scholar"], "title": "Learning Over-Parametrized Two-Layer ReLU Neural Networks beyond NTK", "abstract": "We consider the dynamic of gradient descent for learning a two-layer neural network. We assume the input $x\\in\\mathbb{R}^d$ is drawn from a Gaussian distribution and the label of $x$ satisfies $f^{\\star}(x) = a^{\\top}|W^{\\star}x|$, where $a\\in\\mathbb{R}^d$ is a nonnegative vector and $W^{\\star} \\in\\mathbb{R}^{d\\times d}$ is an orthonormal matrix. We show that an over-parametrized two-layer neural network with ReLU activation, trained by gradient descent from random initialization, can provably learn the ground truth network with population loss at most $o(1/d)$ in polynomial time with polynomial samples. On the other hand, we prove that any kernel method, including Neural Tangent Kernel, with a polynomial number of samples in $d$, has population loss at least $Ω(1 / d)$.", "authors": ["Yuanzhi Li", "Tengyu Ma", "Hongyang R. Zhang"], "categories": ["cs.LG", "math.OC", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-07-09", "url": "https://arxiv.org/abs/2007.04596", "pdf_url": "https://arxiv.org/pdf/2007.04596v1", "arxiv_id": "2007.04596", "doi": null, "citation_count": 29, "influential_citation_count": 3, "has_code": false, "code_url": null, "venue": "Annual Conference Computational Learning Theory", "quality_score": 0.3693}
{"id": "b4405452afa474f9093770b3cfb399a82d29fab30fd2dd4d86a5874fe5a3406f", "sources": ["arxiv", "semantic_scholar"], "title": "Multilevel Graph Matching Networks for Deep Graph Similarity Learning", "abstract": "While the celebrated graph neural networks yield effective representations for individual nodes of a graph, there has been relatively less success in extending to the task of graph similarity learning. Recent work on graph similarity learning has considered either global-level graph-graph interactions or low-level node-node interactions, however ignoring the rich cross-level interactions (e.g., between each node of one graph and the other whole graph). In this paper, we propose a multi-level graph matching network (MGMN) framework for computing the graph similarity between any pair of graph-structured objects in an end-to-end fashion. In particular, the proposed MGMN consists of a node-graph matching network for effectively learning cross-level interactions between each node of one graph and the other whole graph, and a siamese graph neural network to learn global-level interactions between two input graphs. Furthermore, to compensate for the lack of standard benchmark datasets, we have created and collected a set of datasets for both the graph-graph classification and graph-graph regression tasks with different sizes in order to evaluate the effectiveness and robustness of our models. Comprehensive experiments demonstrate that MGMN consistently outperforms state-of-the-art baseline models on both the graph-graph classification and graph-graph regression tasks. Compared with previous work, MGMN also exhibits stronger robustness as the sizes of the two input graphs increase.", "authors": ["Xiang Ling", "Lingfei Wu", "Saizhuo Wang", "Tengfei Ma", "Fangli Xu", "Alex X. Liu", "Chunming Wu", "Shouling Ji"], "categories": ["cs.LG", "cs.AI", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics", "Medicine"], "published_date": "2020-07-08", "url": "https://arxiv.org/abs/2007.04395", "pdf_url": "https://arxiv.org/pdf/2007.04395v4", "arxiv_id": "2007.04395", "doi": "10.1109/TNNLS.2021.3102234", "citation_count": 93, "influential_citation_count": 7, "has_code": false, "code_url": null, "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.4933}
{"id": "d340683bcbee8df81de539bc5fa11a1088274fca6aea118a181fbea5aceadb67", "sources": ["arxiv", "semantic_scholar"], "title": "Dynamic memory to alleviate catastrophic forgetting in continuous learning settings", "abstract": "In medical imaging, technical progress or changes in diagnostic procedures lead to a continuous change in image appearance. Scanner manufacturer, reconstruction kernel, dose, other protocol specific settings or administering of contrast agents are examples that influence image content independent of the scanned biology. Such domain and task shifts limit the applicability of machine learning algorithms in the clinical routine by rendering models obsolete over time. Here, we address the problem of data shifts in a continuous learning scenario by adapting a model to unseen variations in the source domain while counteracting catastrophic forgetting effects. Our method uses a dynamic memory to facilitate rehearsal of a diverse training data subset to mitigate forgetting. We evaluated our approach on routine clinical CT data obtained with two different scanner protocols and synthetic classification tasks. Experiments show that dynamic memory counters catastrophic forgetting in a setting with multiple data shifts without the necessity for explicit knowledge about when these shifts occur.", "authors": ["Johannes Hofmanninger", "Matthias Perkonigg", "James A. Brink", "Oleg Pianykh", "Christian Herold", "Georg Langs"], "categories": ["cs.LG", "cs.CV", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-07-06", "url": "https://arxiv.org/abs/2007.02639", "pdf_url": "https://arxiv.org/pdf/2007.02639v2", "arxiv_id": "2007.02639", "doi": "10.1007/978-3-030-59713-9_35", "citation_count": 30, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": "International Conference on Medical Image Computing and Computer-Assisted Intervention", "quality_score": 0.3728}
{"id": "249b9ca5af007148811adcb7fe10520498b81976d3a74b81c2b3d3aef4a1f90e", "sources": ["arxiv", "semantic_scholar"], "title": "Student-Teacher Curriculum Learning via Reinforcement Learning: Predicting Hospital Inpatient Admission Location", "abstract": "Accurate and reliable prediction of hospital admission location is important due to resource-constraints and space availability in a clinical setting, particularly when dealing with patients who come from the emergency department. In this work we propose a student-teacher network via reinforcement learning to deal with this specific problem. A representation of the weights of the student network is treated as the state and is fed as an input to the teacher network. The teacher network's action is to select the most appropriate batch of data to train the student network on from a training set sorted according to entropy. By validating on three datasets, not only do we show that our approach outperforms state-of-the-art methods on tabular data and performs competitively on image recognition, but also that novel curricula are learned by the teacher network. We demonstrate experimentally that the teacher network can actively learn about the student network and guide it to achieve better performance than if trained alone.", "authors": ["Rasheed el-Bouri", "David Eyre", "Peter Watkinson", "Tingting Zhu", "David Clifton"], "categories": ["cs.LG", "cs.CV", "stat.ML"], "fields_of_study": ["Computer Science", "Psychology", "Mathematics"], "published_date": "2020-07-01", "url": "https://arxiv.org/abs/2007.01135", "pdf_url": "https://arxiv.org/pdf/2007.01135v1", "arxiv_id": "2007.01135", "doi": null, "citation_count": 37, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.3949}
{"id": "6e492de171bb749c5aabc9d0cfa17cf963ce9ee9386354fae60024f2c636ca4f", "sources": ["arxiv", "semantic_scholar"], "title": "Continual Learning: Tackling Catastrophic Forgetting in Deep Neural Networks with Replay Processes", "abstract": "Humans learn all their life long. They accumulate knowledge from a sequence of learning experiences and remember the essential concepts without forgetting what they have learned previously. Artificial neural networks struggle to learn similarly. They often rely on data rigorously preprocessed to learn solutions to specific problems such as classification or regression. In particular, they forget their past learning experiences if trained on new ones. Therefore, artificial neural networks are often inept to deal with real-life settings such as an autonomous-robot that has to learn on-line to adapt to new situations and overcome new problems without forgetting its past learning-experiences. Continual learning (CL) is a branch of machine learning addressing this type of problem. Continual algorithms are designed to accumulate and improve knowledge in a curriculum of learning-experiences without forgetting. In this thesis, we propose to explore continual algorithms with replay processes. Replay processes gather together rehearsal methods and generative replay methods. Generative Replay consists of regenerating past learning experiences with a generative model to remember them. Rehearsal consists of saving a core-set of samples from past learning experiences to rehearse them later. The replay processes make possible a compromise between optimizing the current learning objective and the past ones enabling learning without forgetting in sequences of tasks settings. We show that they are very promising methods for continual learning. Notably, they enable the re-evaluation of past data with new knowledge and the confrontation of data from different learning-experiences. We demonstrate their ability to learn continually through unsupervised learning, supervised learning and reinforcement learning tasks.", "authors": ["Timothée Lesort"], "categories": ["cs.LG", "cs.AI", "cs.NE"], "fields_of_study": ["Computer Science"], "published_date": "2020-07-01", "url": "https://arxiv.org/abs/2007.00487", "pdf_url": "https://arxiv.org/pdf/2007.00487v3", "arxiv_id": "2007.00487", "doi": null, "citation_count": 27, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3618}
{"id": "fe7c01f37e47d97bed2d30b42e3a69d44c195dc7b09df0f52b532d3dc52bf791", "sources": ["arxiv", "semantic_scholar"], "title": "Continual Learning from the Perspective of Compression", "abstract": "Connectionist models such as neural networks suffer from catastrophic forgetting. In this work, we study this problem from the perspective of information theory and define forgetting as the increase of description lengths of previous data when they are compressed with a sequentially learned model. In addition, we show that continual learning approaches based on variational posterior approximation and generative replay can be considered as approximations to two prequential coding methods in compression, namely, the Bayesian mixture code and maximum likelihood (ML) plug-in code. We compare these approaches in terms of both compression and forgetting and empirically study the reasons that limit the performance of continual learning methods based on variational posterior approximation. To address these limitations, we propose a new continual learning method that combines ML plug-in and Bayesian mixture codes.", "authors": ["Xu He", "Min Lin"], "categories": ["cs.LG", "cs.NE", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-06-26", "url": "https://arxiv.org/abs/2006.15078", "pdf_url": "https://arxiv.org/pdf/2006.15078v1", "arxiv_id": "2006.15078", "doi": null, "citation_count": 3, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1505}
{"id": "5a499cd929c8e9a9e7e914e0436fa20edc9c02574134b445c6c9e291e152cbe2", "sources": ["arxiv", "semantic_scholar"], "title": "Storing Encoded Episodes as Concepts for Continual Learning", "abstract": "The two main challenges faced by continual learning approaches are catastrophic forgetting and memory limitations on the storage of data. To cope with these challenges, we propose a novel, cognitively-inspired approach which trains autoencoders with Neural Style Transfer to encode and store images. Reconstructed images from encoded episodes are replayed when training the classifier model on a new task to avoid catastrophic forgetting. The loss function for the reconstructed images is weighted to reduce its effect during classifier training to cope with image degradation. When the system runs out of memory the encoded episodes are converted into centroids and covariance matrices, which are used to generate pseudo-images during classifier training, keeping classifier performance stable with less memory. Our approach increases classification accuracy by 13-17% over state-of-the-art methods on benchmark datasets, while requiring 78% less storage space.", "authors": ["Ali Ayub", "Alan R. Wagner"], "categories": ["cs.CV", "cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-06-26", "url": "https://arxiv.org/abs/2007.06637", "pdf_url": "https://arxiv.org/pdf/2007.06637v1", "arxiv_id": "2007.06637", "doi": null, "citation_count": 13, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2865}
{"id": "8eb8d2edb93abca0f145e228afd5c1717301d54276756a5f670286e3cb7cd301", "sources": ["arxiv", "semantic_scholar"], "title": "Lifelong Learning of Graph Neural Networks for Open-World Node Classification", "abstract": "Graph neural networks (GNNs) have emerged as the standard method for numerous tasks on graph-structured data such as node classification. However, real-world graphs are often evolving over time and even new classes may arise. We model these challenges as an instance of lifelong learning, in which a learner faces a sequence of tasks and may take over knowledge acquired in past tasks. Such knowledge may be stored explicitly as historic data or implicitly within model parameters. In this work, we systematically analyze the influence of implicit and explicit knowledge. Therefore, we present an incremental training method for lifelong learning on graphs and introduce a new measure based on $k$-neighborhood time differences to address variances in the historic data. We apply our training method to five representative GNN architectures and evaluate them on three new lifelong node classification datasets. Our results show that no more than 50% of the GNN's receptive field is necessary to retain at least 95% accuracy compared to training over the complete history of the graph data. Furthermore, our experiments confirm that implicit knowledge becomes more important when fewer explicit knowledge is available.", "authors": ["Lukas Galke", "Benedikt Franke", "Tobias Zielke", "Ansgar Scherp"], "categories": ["cs.LG", "cs.SI", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-06-25", "url": "https://arxiv.org/abs/2006.14422", "pdf_url": "https://arxiv.org/pdf/2006.14422v4", "arxiv_id": "2006.14422", "doi": "10.1109/IJCNN52387.2021.9533412", "citation_count": 44, "influential_citation_count": 5, "has_code": false, "code_url": null, "venue": "IEEE International Joint Conference on Neural Network", "quality_score": 0.4133}
{"id": "03a10306fdc3475416771418fe003a2ee50f0450e53871947128d18317484e55", "sources": ["arxiv", "semantic_scholar"], "title": "DOME: Recommendations for supervised machine learning validation in biology", "abstract": "Modern biology frequently relies on machine learning to provide predictions and improve decision processes. There have been recent calls for more scrutiny on machine learning performance and possible limitations. Here we present a set of community-wide recommendations aiming to help establish standards of supervised machine learning validation in biology. Adopting a structured methods description for machine learning based on data, optimization, model, evaluation (DOME) will aim to help both reviewers and readers to better understand and assess the performance and limitations of a method or outcome. The recommendations are formulated as questions to anyone wishing to pursue implementation of a machine learning algorithm. Answers to these questions can be easily included in the supplementary material of published papers.", "authors": ["Ian Walsh", "Dmytro Fishman", "Dario Garcia-Gasulla", "Tiina Titma", "Gianluca Pollastri", "The ELIXIR Machine Learning focus group", "Jen Harrow", "Fotis E. Psomopoulos", "Silvio C. E. Tosatto"], "categories": ["q-bio.OT", "cs.LG"], "fields_of_study": ["Biology", "Computer Science"], "published_date": "2020-06-25", "url": "https://arxiv.org/abs/2006.16189", "pdf_url": "https://arxiv.org/pdf/2006.16189v4", "arxiv_id": "2006.16189", "doi": null, "citation_count": 4, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1747}
{"id": "0b27b567af2fd6b6873f529f894b334d67ea6de6d26c8845c3988cc200a3d7a7", "sources": ["arxiv", "semantic_scholar"], "title": "The Gaussian equivalence of generative models for learning with shallow neural networks", "abstract": "Understanding the impact of data structure on the computational tractability of learning is a key challenge for the theory of neural networks. Many theoretical works do not explicitly model training data, or assume that inputs are drawn component-wise independently from some simple probability distribution. Here, we go beyond this simple paradigm by studying the performance of neural networks trained on data drawn from pre-trained generative models. This is possible due to a Gaussian equivalence stating that the key metrics of interest, such as the training and test errors, can be fully captured by an appropriately chosen Gaussian model. We provide three strands of rigorous, analytical and numerical evidence corroborating this equivalence. First, we establish rigorous conditions for the Gaussian equivalence to hold in the case of single-layer generative models, as well as deterministic rates for convergence in distribution. Second, we leverage this equivalence to derive a closed set of equations describing the generalisation performance of two widely studied machine learning problems: two-layer neural networks trained using one-pass stochastic gradient descent, and full-batch pre-learned features or kernel methods. Finally, we perform experiments demonstrating how our theory applies to deep, pre-trained generative models. These results open a viable path to the theoretical study of machine learning models with realistic data.", "authors": ["Sebastian Goldt", "Bruno Loureiro", "Galen Reeves", "Florent Krzakala", "Marc Mézard", "Lenka Zdeborová"], "categories": ["stat.ML", "cond-mat.dis-nn", "cond-mat.stat-mech", "cs.LG"], "fields_of_study": ["Computer Science", "Mathematics", "Physics"], "published_date": "2020-06-25", "url": "https://arxiv.org/abs/2006.14709", "pdf_url": "https://arxiv.org/pdf/2006.14709v3", "arxiv_id": "2006.14709", "doi": null, "citation_count": 130, "influential_citation_count": 5, "has_code": true, "code_url": "https://github.com/sgoldt/gaussian-equiv-2layer", "venue": "Mathematical and Scientific Machine Learning", "quality_score": 0.5293}
{"id": "034866301046820e37753f1b64878d52e85f95fde2068e3223e5d210ac36b7ee", "sources": ["arxiv", "semantic_scholar"], "title": "Optimization and Generalization of Regularization-Based Continual Learning: a Loss Approximation Viewpoint", "abstract": "Neural networks have achieved remarkable success in many cognitive tasks. However, when they are trained sequentially on multiple tasks without access to old data, their performance on early tasks tend to drop significantly. This problem is often referred to as catastrophic forgetting, a key challenge in continual learning of neural networks. The regularization-based approach is one of the primary classes of methods to alleviate catastrophic forgetting. In this paper, we provide a novel viewpoint of regularization-based continual learning by formulating it as a second-order Taylor approximation of the loss function of each task. This viewpoint leads to a unified framework that can be instantiated to derive many existing algorithms such as Elastic Weight Consolidation and Kronecker factored Laplace approximation. Based on this viewpoint, we study the optimization aspects (i.e., convergence) as well as generalization properties (i.e., finite-sample guarantees) of regularization-based continual learning. Our theoretical results indicate the importance of accurate approximation of the Hessian matrix. The experimental results on several benchmarks provide empirical validation of our theoretical findings.", "authors": ["Dong Yin", "Mehrdad Farajtabar", "Ang Li", "Nir Levine", "Alex Mott"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-06-19", "url": "https://arxiv.org/abs/2006.10974", "pdf_url": "https://arxiv.org/pdf/2006.10974v3", "arxiv_id": "2006.10974", "doi": null, "citation_count": 20, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3306}
{"id": "f5621cb0e86f432417cf39e2fe7573addc8eaf9b4d06a4ca67694f4caf8b7bbd", "sources": ["arxiv", "semantic_scholar"], "title": "Neural Topic Modeling with Continual Lifelong Learning", "abstract": "Lifelong learning has recently attracted attention in building machine learning systems that continually accumulate and transfer knowledge to help future learning. Unsupervised topic modeling has been popularly used to discover topics from document collections. However, the application of topic modeling is challenging due to data sparsity, e.g., in a small collection of (short) documents and thus, generate incoherent topics and sub-optimal document representations. To address the problem, we propose a lifelong learning framework for neural topic modeling that can continuously process streams of document collections, accumulate topics and guide future topic modeling tasks by knowledge transfer from several sources to better deal with the sparse data. In the lifelong process, we particularly investigate jointly: (1) sharing generative homologies (latent topics) over lifetime to transfer prior knowledge, and (2) minimizing catastrophic forgetting to retain the past learning via novel selective data augmentation, co-training and topic regularization approaches. Given a stream of document collections, we apply the proposed Lifelong Neural Topic Modeling (LNTM) framework in modeling three sparse document collections as future tasks and demonstrate improved performance quantified by perplexity, topic coherence and information retrieval task.", "authors": ["Pankaj Gupta", "Yatin Chaudhary", "Thomas Runkler", "Hinrich Schütze"], "categories": ["cs.CL", "cs.IR", "cs.LG", "cs.NE"], "fields_of_study": ["Computer Science"], "published_date": "2020-06-19", "url": "https://arxiv.org/abs/2006.10909", "pdf_url": "https://arxiv.org/pdf/2006.10909v2", "arxiv_id": "2006.10909", "doi": null, "citation_count": 55, "influential_citation_count": 6, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.437}
{"id": "b61cda0cfb668c210dd82abfd630ecc924b0bedcf3a93b4b717591a306a8cc48", "sources": ["arxiv", "semantic_scholar"], "title": "A benchmark study on reliable molecular supervised learning via Bayesian learning", "abstract": "Virtual screening aims to find desirable compounds from chemical library by using computational methods. For this purpose with machine learning, model outputs that can be interpreted as predictive probability will be beneficial, in that a high prediction score corresponds to high probability of correctness. In this work, we present a study on the prediction performance and reliability of graph neural networks trained with the recently proposed Bayesian learning algorithms. Our work shows that Bayesian learning algorithms allow well-calibrated predictions for various GNN architectures and classification tasks. Also, we show the implications of reliable predictions on virtual screening, where Bayesian learning may lead to higher success in finding hit compounds.", "authors": ["Doyeong Hwang", "Grace Lee", "Hanseok Jo", "Seyoul Yoon", "Seongok Ryu"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-06-12", "url": "https://arxiv.org/abs/2006.07021", "pdf_url": "https://arxiv.org/pdf/2006.07021v2", "arxiv_id": "2006.07021", "doi": null, "citation_count": 9, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.25}
{"id": "3218f77700d40d14044a9f80e3e877ca5b5cd34e6a2900650d91f3d0949fe4c7", "sources": ["arxiv", "semantic_scholar"], "title": "Anti-Transfer Learning for Task Invariance in Convolutional Neural Networks for Speech Processing", "abstract": "We introduce the novel concept of anti-transfer learning for speech processing with convolutional neural networks. While transfer learning assumes that the learning process for a target task will benefit from re-using representations learned for another task, anti-transfer avoids the learning of representations that have been learned for an orthogonal task, i.e., one that is not relevant and potentially misleading for the target task, such as speaker identity for speech recognition or speech content for emotion recognition. In anti-transfer learning, we penalize similarity between activations of a network being trained and another one previously trained on an orthogonal task, which yields more suitable representations. This leads to better generalization and provides a degree of control over correlations that are spurious or undesirable, e.g. to avoid social bias. We have implemented anti-transfer for convolutional neural networks in different configurations with several similarity metrics and aggregation functions, which we evaluate and analyze with several speech and audio tasks and settings, using six datasets. We show that anti-transfer actually leads to the intended invariance to the orthogonal task and to more appropriate features for the target task at hand. Anti-transfer learning consistently improves classification accuracy in all test cases. While anti-transfer creates computation and memory cost at training time, there is relatively little computation cost when using pre-trained models for orthogonal tasks. Anti-transfer is widely applicable and particularly useful where a specific invariance is desirable or where trained models are available and labeled data for orthogonal tasks are difficult to obtain.", "authors": ["Eric Guizzo", "Tillman Weyde", "Giacomo Tarroni"], "categories": ["cs.LG", "cs.NE", "cs.SD", "eess.AS", "stat.ML"], "fields_of_study": ["Computer Science", "Engineering", "Mathematics"], "published_date": "2020-06-11", "url": "https://arxiv.org/abs/2006.06494", "pdf_url": "https://arxiv.org/pdf/2006.06494v2", "arxiv_id": "2006.06494", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.0}
{"id": "97037199fedc58853291541b4a52f3f1f3588c5684896ce0551a250b3b4b5744", "sources": ["arxiv", "semantic_scholar"], "title": "Self-Supervised Learning Aided Class-Incremental Lifelong Learning", "abstract": "Lifelong or continual learning remains to be a challenge for artificial neural network, as it is required to be both stable for preservation of old knowledge and plastic for acquisition of new knowledge. It is common to see previous experience get overwritten, which leads to the well-known issue of catastrophic forgetting, especially in the scenario of class-incremental learning (Class-IL). Recently, many lifelong learning methods have been proposed to avoid catastrophic forgetting. However, models which learn without replay of the input data, would encounter another problem which has been ignored, and we refer to it as prior information loss (PIL). In training procedure of Class-IL, as the model has no knowledge about following tasks, it would only extract features necessary for tasks learned so far, whose information is insufficient for joint classification. In this paper, our empirical results on several image datasets show that PIL limits the performance of current state-of-the-art method for Class-IL, the orthogonal weights modification (OWM) algorithm. Furthermore, we propose to combine self-supervised learning, which can provide effective representations without requiring labels, with Class-IL to partly get around this problem. Experiments show superiority of proposed method to OWM, as well as other strong baselines.", "authors": ["Song Zhang", "Gehui Shen", "Jinsong Huang", "Zhi-Hong Deng"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-06-10", "url": "https://arxiv.org/abs/2006.05882", "pdf_url": "https://arxiv.org/pdf/2006.05882v4", "arxiv_id": "2006.05882", "doi": null, "citation_count": 14, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.294}
{"id": "ef90e79548aa934c5d72a62501ebd2b930ed533f01eeb36766b21e48404afbec", "sources": ["arxiv", "semantic_scholar"], "title": "Variational Auto-Regressive Gaussian Processes for Continual Learning", "abstract": "Through sequential construction of posteriors on observing data online, Bayes' theorem provides a natural framework for continual learning. We develop Variational Auto-Regressive Gaussian Processes (VAR-GPs), a principled posterior updating mechanism to solve sequential tasks in continual learning. By relying on sparse inducing point approximations for scalable posteriors, we propose a novel auto-regressive variational distribution which reveals two fruitful connections to existing results in Bayesian inference, expectation propagation and orthogonal inducing points. Mean predictive entropy estimates show VAR-GPs prevent catastrophic forgetting, which is empirically supported by strong performance on modern continual learning benchmarks against competitive baselines. A thorough ablation study demonstrates the efficacy of our modeling choices.", "authors": ["Sanyam Kapoor", "Theofanis Karaletsos", "Thang D. Bui"], "categories": ["stat.ML", "cs.LG"], "fields_of_study": ["Mathematics", "Computer Science"], "published_date": "2020-06-09", "url": "https://arxiv.org/abs/2006.05468", "pdf_url": "https://arxiv.org/pdf/2006.05468v3", "arxiv_id": "2006.05468", "doi": null, "citation_count": 32, "influential_citation_count": 3, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.3796}
{"id": "0a2d750bab94e8eafe7fb705b211d371d3405ec1bd53db617cfb86e38764c3ce", "sources": ["arxiv", "semantic_scholar"], "title": "Optimal Continual Learning has Perfect Memory and is NP-hard", "abstract": "Continual Learning (CL) algorithms incrementally learn a predictor or representation across multiple sequentially observed tasks. Designing CL algorithms that perform reliably and avoid so-called catastrophic forgetting has proven a persistent challenge. The current paper develops a theoretical approach that explains why. In particular, we derive the computational properties which CL algorithms would have to possess in order to avoid catastrophic forgetting. Our main finding is that such optimal CL algorithms generally solve an NP-hard problem and will require perfect memory to do so. The findings are of theoretical interest, but also explain the excellent performance of CL algorithms using experience replay, episodic memory and core sets relative to regularization-based approaches.", "authors": ["Jeremias Knoblauch", "Hisham Husain", "Tom Diethe"], "categories": ["cs.LG", "cs.AI", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-06-09", "url": "https://arxiv.org/abs/2006.05188", "pdf_url": "https://arxiv.org/pdf/2006.05188v1", "arxiv_id": "2006.05188", "doi": null, "citation_count": 117, "influential_citation_count": 8, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.518}
{"id": "a46df58443da04d28f620ab1da4f327037aa6c4f9df29becd4a19ead2d886371", "sources": ["arxiv", "semantic_scholar"], "title": "Learning to Stop While Learning to Predict", "abstract": "There is a recent surge of interest in designing deep architectures based on the update steps in traditional algorithms, or learning neural networks to improve and replace traditional algorithms. While traditional algorithms have certain stopping criteria for outputting results at different iterations, many algorithm-inspired deep models are restricted to a ``fixed-depth'' for all inputs. Similar to algorithms, the optimal depth of a deep architecture may be different for different input instances, either to avoid ``over-thinking'', or because we want to compute less for operations converged already. In this paper, we tackle this varying depth problem using a steerable architecture, where a feed-forward deep model and a variational stopping policy are learned together to sequentially determine the optimal number of layers for each input instance. Training such architecture is very challenging. We provide a variational Bayes perspective and design a novel and effective training procedure which decomposes the task into an oracle model learning stage and an imitation stage. Experimentally, we show that the learned deep model along with the stopping policy improves the performances on a diverse set of tasks, including learning sparse recovery, few-shot meta learning, and computer vision tasks.", "authors": ["Xinshi Chen", "Hanjun Dai", "Yu Li", "Xin Gao", "Le Song"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-06-09", "url": "https://arxiv.org/abs/2006.05082", "pdf_url": "https://arxiv.org/pdf/2006.05082v1", "arxiv_id": "2006.05082", "doi": null, "citation_count": 60, "influential_citation_count": 5, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.4463}
{"id": "7603beb7b339652313efaa570b60b073448a563820daa6bf2c2c63bf8c0c391a", "sources": ["arxiv", "semantic_scholar"], "title": "CiwGAN and fiwGAN: Encoding information in acoustic data to model lexical learning with Generative Adversarial Networks", "abstract": "How can deep neural networks encode information that corresponds to words in human speech into raw acoustic data? This paper proposes two neural network architectures for modeling unsupervised lexical learning from raw acoustic inputs, ciwGAN (Categorical InfoWaveGAN) and fiwGAN (Featural InfoWaveGAN), that combine a Deep Convolutional GAN architecture for audio data (WaveGAN; arXiv:1705.07904) with an information theoretic extension of GAN -- InfoGAN (arXiv:1606.03657), and propose a new latent space structure that can model featural learning simultaneously with a higher level classification and allows for a very low-dimension vector representation of lexical items. Lexical learning is modeled as emergent from an architecture that forces a deep neural network to output data such that unique information is retrievable from its acoustic outputs. The networks trained on lexical items from TIMIT learn to encode unique information corresponding to lexical items in the form of categorical variables in their latent space. By manipulating these variables, the network outputs specific lexical items. The network occasionally outputs innovative lexical items that violate training data, but are linguistically interpretable and highly informative for cognitive modeling and neural network interpretability. Innovative outputs suggest that phonetic and phonological representations learned by the network can be productively recombined and directly paralleled to productivity in human speech: a fiwGAN network trained on `suit' and `dark' outputs innovative `start', even though it never saw `start' or even a [st] sequence in the training data. We also argue that setting latent featural codes to values well beyond training range results in almost categorical generation of prototypical lexical items and reveals underlying values of each latent code.", "authors": ["Gašper Beguš"], "categories": ["cs.CL", "cs.LG"], "fields_of_study": ["Medicine", "Computer Science"], "published_date": "2020-06-04", "url": "https://arxiv.org/abs/2006.02951", "pdf_url": "https://arxiv.org/pdf/2006.02951v3", "arxiv_id": "2006.02951", "doi": "10.1016/j.neunet.2021.03.017", "citation_count": 40, "influential_citation_count": 6, "has_code": false, "code_url": null, "venue": "Neural Networks", "quality_score": 0.4225}
{"id": "562399ffa9251e26d15aac0e5b8627436ddaeb8f831be891f950b66d532f3216", "sources": ["arxiv", "semantic_scholar"], "title": "Network Comparison with Interpretable Contrastive Network Representation Learning", "abstract": "Identifying unique characteristics in a network through comparison with another network is an essential network analysis task. For example, with networks of protein interactions obtained from normal and cancer tissues, we can discover unique types of interactions in cancer tissues. This analysis task could be greatly assisted by contrastive learning, which is an emerging analysis approach to discover salient patterns in one dataset relative to another. However, existing contrastive learning methods cannot be directly applied to networks as they are designed only for high-dimensional data analysis. To address this problem, we introduce a new analysis approach called contrastive network representation learning (cNRL). By integrating two machine learning schemes, network representation learning and contrastive learning, cNRL enables embedding of network nodes into a low-dimensional representation that reveals the uniqueness of one network compared to another. Within this approach, we also design a method, named i-cNRL, which offers interpretability in the learned results, allowing for understanding which specific patterns are only found in one network. We demonstrate the effectiveness of i-cNRL for network comparison with multiple network models and real-world datasets. Furthermore, we compare i-cNRL and other potential cNRL algorithm designs through quantitative and qualitative evaluations.", "authors": ["Takanori Fujiwara", "Jian Zhao", "Francine Chen", "Yaoliang Yu", "Kwan-Liu Ma"], "categories": ["cs.LG", "cs.SI", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics", "Medicine"], "published_date": "2020-05-25", "url": "https://arxiv.org/abs/2005.12419", "pdf_url": "https://arxiv.org/pdf/2005.12419v2", "arxiv_id": "2005.12419", "doi": "10.52933/jdssv.v2i5.56", "citation_count": 7, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Journal of Data Science Statistics and Visualisation", "quality_score": 0.2258}
{"id": "dd55ba4c5ac230d3463dab05b32295ca827b57dddc5d7537463f5e4293d6db1e", "sources": ["arxiv", "semantic_scholar"], "title": "Modularizing Deep Learning via Pairwise Learning With Kernels", "abstract": "By redefining the conventional notions of layers, we present an alternative view on finitely wide, fully trainable deep neural networks as stacked linear models in feature spaces, leading to a kernel machine interpretation. Based on this construction, we then propose a provably optimal modular learning framework for classification that does not require between-module backpropagation. This modular approach brings new insights into the label requirement of deep learning: It leverages only implicit pairwise labels (weak supervision) when learning the hidden modules. When training the output module, on the other hand, it requires full supervision but achieves high label efficiency, needing as few as 10 randomly selected labeled examples (one from each class) to achieve 94.88% accuracy on CIFAR-10 using a ResNet-18 backbone. Moreover, modular training enables fully modularized deep learning workflows, which then simplify the design and implementation of pipelines and improve the maintainability and reusability of models. To showcase the advantages of such a modularized workflow, we describe a simple yet reliable method for estimating reusability of pre-trained modules as well as task transferability in a transfer learning setting. At practically no computation overhead, it precisely described the task space structure of 15 binary classification tasks from CIFAR-10.", "authors": ["Shiyu Duan", "Shujian Yu", "Jose Principe"], "categories": ["stat.ML", "cs.LG"], "fields_of_study": ["Mathematics", "Computer Science", "Medicine"], "published_date": "2020-05-12", "url": "https://arxiv.org/abs/2005.05541", "pdf_url": "https://arxiv.org/pdf/2005.05541v2", "arxiv_id": "2005.05541", "doi": "10.1109/TNNLS.2020.3042346", "citation_count": 22, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.3404}
{"id": "7bf18016ea64827057146323c15cfe52c0004aab74f8852e6abee6feccba37f0", "sources": ["arxiv", "semantic_scholar"], "title": "Visually Impaired Aid using Convolutional Neural Networks, Transfer Learning, and Particle Competition and Cooperation", "abstract": "Navigation and mobility are some of the major problems faced by visually impaired people in their daily lives. Advances in computer vision led to the proposal of some navigation systems. However, most of them require expensive and/or heavy hardware. In this paper we propose the use of convolutional neural networks (CNN), transfer learning, and semi-supervised learning (SSL) to build a framework aimed at the visually impaired aid. It has low computational costs and, therefore, may be implemented on current smartphones, without relying on any additional equipment. The smartphone camera can be used to automatically take pictures of the path ahead. Then, they will be immediately classified, providing almost instantaneous feedback to the user. We also propose a dataset to train the classifiers, including indoor and outdoor situations with different types of light, floor, and obstacles. Many different CNN architectures are evaluated as feature extractors and classifiers, by fine-tuning weights pre-trained on a much larger dataset. The graph-based SSL method, known as particle competition and cooperation, is also used for classification, allowing feedback from the user to be incorporated without retraining the underlying network. 92\\% and 80\\% classification accuracy is achieved in the proposed dataset in the best supervised and SSL scenarios, respectively.", "authors": ["Fabricio Breve", "Carlos Norberto Fischer"], "categories": ["cs.CV", "cs.LG", "cs.NE"], "fields_of_study": ["Computer Science"], "published_date": "2020-05-09", "url": "https://arxiv.org/abs/2005.04473", "pdf_url": "https://arxiv.org/pdf/2005.04473v1", "arxiv_id": "2005.04473", "doi": "10.1109/IJCNN48605.2020.9207606", "citation_count": 5, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "IEEE International Joint Conference on Neural Network", "quality_score": 0.1945}
{"id": "63b432a23fc87c90275310a4fc74655ede8cc1749aee8cc114ea83437ce9173f", "sources": ["arxiv", "semantic_scholar"], "title": "Continual Learning Using Multi-view Task Conditional Neural Networks", "abstract": "Conventional deep learning models have limited capacity in learning multiple tasks sequentially. The issue of forgetting the previously learned tasks in continual learning is known as catastrophic forgetting or interference. When the input data or the goal of learning change, a continual model will learn and adapt to the new status. However, the model will not remember or recognise any revisits to the previous states. This causes performance reduction and re-training curves in dealing with periodic or irregularly reoccurring changes in the data or goals. The changes in goals or data are referred to as new tasks in a continual learning model. Most of the continual learning methods have a task-known setup in which the task identities are known in advance to the learning model. We propose Multi-view Task Conditional Neural Networks (Mv-TCNN) that does not require to known the reoccurring tasks in advance. We evaluate our model on standard datasets using MNIST, CIFAR10, CIFAR100, and also a real-world dataset that we have collected in a remote healthcare monitoring study (i.e. TIHM dataset). The proposed model outperforms the state-of-the-art solutions in continual learning and adapting to new tasks that are not defined in advance.", "authors": ["Honglin Li", "Payam Barnaghi", "Shirin Enshaeifar", "Frieder Ganz"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-05-08", "url": "https://arxiv.org/abs/2005.05080", "pdf_url": "https://arxiv.org/pdf/2005.05080v3", "arxiv_id": "2005.05080", "doi": null, "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.0753}
{"id": "e1e826fded41bcb25f3fa2fbac57706a77452448ac334ad9c271009fd36256e6", "sources": ["arxiv", "semantic_scholar"], "title": "On Interpretability of Deep Learning based Skin Lesion Classifiers using Concept Activation Vectors", "abstract": "Deep learning based medical image classifiers have shown remarkable prowess in various application areas like ophthalmology, dermatology, pathology, and radiology. However, the acceptance of these Computer-Aided Diagnosis (CAD) systems in real clinical setups is severely limited primarily because their decision-making process remains largely obscure. This work aims at elucidating a deep learning based medical image classifier by verifying that the model learns and utilizes similar disease-related concepts as described and employed by dermatologists. We used a well-trained and high performing neural network developed by REasoning for COmplex Data (RECOD) Lab for classification of three skin tumours, i.e. Melanocytic Naevi, Melanoma and Seborrheic Keratosis and performed a detailed analysis on its latent space. Two well established and publicly available skin disease datasets, PH2 and derm7pt, are used for experimentation. Human understandable concepts are mapped to RECOD image classification model with the help of Concept Activation Vectors (CAVs), introducing a novel training and significance testing paradigm for CAVs. Our results on an independent evaluation set clearly shows that the classifier learns and encodes human understandable concepts in its latent representation. Additionally, TCAV scores (Testing with CAVs) suggest that the neural network indeed makes use of disease-related concepts in the correct way when making predictions. We anticipate that this work can not only increase confidence of medical practitioners on CAD but also serve as a stepping stone for further development of CAV-based neural network interpretation methods.", "authors": ["Adriano Lucieri", "Muhammad Naseer Bajwa", "Stephan Alexander Braun", "Muhammad Imran Malik", "Andreas Dengel", "Sheraz Ahmed"], "categories": ["cs.LG", "cs.CV", "eess.IV", "stat.ML"], "fields_of_study": ["Computer Science", "Engineering", "Mathematics"], "published_date": "2020-05-05", "url": "https://arxiv.org/abs/2005.02000", "pdf_url": "https://arxiv.org/pdf/2005.02000v1", "arxiv_id": "2005.02000", "doi": "10.1109/IJCNN48605.2020.9206946", "citation_count": 77, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "IEEE International Joint Conference on Neural Network", "quality_score": 0.473}
{"id": "1800639abb14b3b861029133a71df5007ff447a7990b336557a02360d51cfd9b", "sources": ["arxiv", "semantic_scholar"], "title": "Explaining How Deep Neural Networks Forget by Deep Visualization", "abstract": "Explaining the behaviors of deep neural networks, usually considered as black boxes, is critical especially when they are now being adopted over diverse aspects of human life. Taking the advantages of interpretable machine learning (interpretable ML), this paper proposes a novel tool called Catastrophic Forgetting Dissector (or CFD) to explain catastrophic forgetting in continual learning settings. We also introduce a new method called Critical Freezing based on the observations of our tool. Experiments on ResNet articulate how catastrophic forgetting happens, particularly showing which components of this famous network are forgetting. Our new continual learning algorithm defeats various recent techniques by a significant margin, proving the capability of the investigation. Critical freezing not only attacks catastrophic forgetting but also exposes explainability.", "authors": ["Giang Nguyen", "Shuan Chen", "Tae Joon Jun", "Daeyoung Kim"], "categories": ["cs.LG", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2020-05-03", "url": "https://arxiv.org/abs/2005.01004", "pdf_url": "https://arxiv.org/pdf/2005.01004v3", "arxiv_id": "2005.01004", "doi": "10.1007/978-3-030-68796-0_12", "citation_count": 10, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "ICPR 2020", "quality_score": 0.2603}
{"id": "edd88c32200816b741e897ac3a8a960a1645985183ccba41eae3a32075d36cf8", "sources": ["arxiv", "semantic_scholar"], "title": "Reducing catastrophic forgetting with learning on synthetic data", "abstract": "Catastrophic forgetting is a problem caused by neural networks' inability to learn data in sequence. After learning two tasks in sequence, performance on the first one drops significantly. This is a serious disadvantage that prevents many deep learning applications to real-life problems where not all object classes are known beforehand; or change in data requires adjustments to the model. To reduce this problem we investigate the use of synthetic data, namely we answer a question: Is it possible to generate such data synthetically which learned in sequence does not result in catastrophic forgetting? We propose a method to generate such data in two-step optimisation process via meta-gradients. Our experimental results on Split-MNIST dataset show that training a model on such synthetic data in sequence does not result in catastrophic forgetting. We also show that our method of generating data is robust to different learning scenarios.", "authors": ["Wojciech Masarczyk", "Ivona Tautkute"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-04-29", "url": "https://arxiv.org/abs/2004.14046", "pdf_url": "https://arxiv.org/pdf/2004.14046v1", "arxiv_id": "2004.14046", "doi": "10.1109/CVPRW50498.2020.00134", "citation_count": 40, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.4032}
{"id": "108de27e88e67cdd3ea330f2c9a45207cdc7336132a9cb66dfa9e0939b34ae5d", "sources": ["arxiv", "semantic_scholar"], "title": "Natural Way to Overcome the Catastrophic Forgetting in Neural Networks", "abstract": "Not so long ago, a method was discovered that successfully overcomes the catastrophic forgetting in neural networks. Although we know about the cases of using this method to preserve skills when adapting pre-trained networks to particular tasks, it has not obtained widespread distribution yet. In this paper, we would like to propose an alternative method of overcoming catastrophic forgetting based on the total absolute signal passing through each connection in the network. This method has a simple implementation and seems to us essentially close to the processes occurring in the brain of animals to preserve previously learned skills during subsequent learning. We hope that the ease of implementation of this method will serve its wide application.", "authors": ["Alexey Kutalev"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-04-27", "url": "https://arxiv.org/abs/2005.07107", "pdf_url": "https://arxiv.org/pdf/2005.07107v2", "arxiv_id": "2005.07107", "doi": "10.25559/SITITO.16.202002.331-337", "citation_count": 8, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2386}
{"id": "d775d91d90b2fc0e3903d425f252cb0c460a74653fb33883a13fe67306882bb2", "sources": ["arxiv", "semantic_scholar"], "title": "Learning Constrained Dynamics with Gauss Principle adhering Gaussian Processes", "abstract": "The identification of the constrained dynamics of mechanical systems is often challenging. Learning methods promise to ease an analytical analysis, but require considerable amounts of data for training. We propose to combine insights from analytical mechanics with Gaussian process regression to improve the model's data efficiency and constraint integrity. The result is a Gaussian process model that incorporates a priori constraint knowledge such that its predictions adhere to Gauss' principle of least constraint. In return, predictions of the system's acceleration naturally respect potentially non-ideal (non-)holonomic equality constraints. As corollary results, our model enables to infer the acceleration of the unconstrained system from data of the constrained system and enables knowledge transfer between differing constraint configurations.", "authors": ["A. Rene Geist", "Sebastian Trimpe"], "categories": ["cs.LG", "cs.RO", "eess.SY", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics", "Engineering"], "published_date": "2020-04-23", "url": "https://arxiv.org/abs/2004.11238", "pdf_url": "https://arxiv.org/pdf/2004.11238v1", "arxiv_id": "2004.11238", "doi": null, "citation_count": 23, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Conference on Learning for Dynamics & Control", "quality_score": 0.3451}
{"id": "5c109edb784ac28720cba9c282433e8d4d54f0ecaed6a02c7a2a98e1ff2b32b8", "sources": ["arxiv", "semantic_scholar"], "title": "Continual Learning of Object Instances", "abstract": "We propose continual instance learning - a method that applies the concept of continual learning to the task of distinguishing instances of the same object category. We specifically focus on the car object, and incrementally learn to distinguish car instances from each other with metric learning. We begin our paper by evaluating current techniques. Establishing that catastrophic forgetting is evident in existing methods, we then propose two remedies. Firstly, we regularise metric learning via Normalised Cross-Entropy. Secondly, we augment existing models with synthetic data transfer. Our extensive experiments on three large-scale datasets, using two different architectures for five different continual learning methods, reveal that Normalised cross-entropy and synthetic transfer leads to less forgetting in existing techniques.", "authors": ["Kishan Parshotam", "Mert Kilickaya"], "categories": ["cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2020-04-22", "url": "https://arxiv.org/abs/2004.10862", "pdf_url": "https://arxiv.org/pdf/2004.10862v1", "arxiv_id": "2004.10862", "doi": "10.1109/CVPRW50498.2020.00120", "citation_count": 4, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.1747}
{"id": "dc06ba3845760ca72c5deecbcecc18e735ff082ea7be4d6df1a213aec943d24c", "sources": ["arxiv", "semantic_scholar"], "title": "Generative Feature Replay For Class-Incremental Learning", "abstract": "Humans are capable of learning new tasks without forgetting previous ones, while neural networks fail due to catastrophic forgetting between new and previously-learned tasks. We consider a class-incremental setting which means that the task-ID is unknown at inference time. The imbalance between old and new classes typically results in a bias of the network towards the newest ones. This imbalance problem can either be addressed by storing exemplars from previous tasks, or by using image replay methods. However, the latter can only be applied to toy datasets since image generation for complex datasets is a hard problem. We propose a solution to the imbalance problem based on generative feature replay which does not require any exemplars. To do this, we split the network into two parts: a feature extractor and a classifier. To prevent forgetting, we combine generative feature replay in the classifier with feature distillation in the feature extractor. Through feature generation, our method reduces the complexity of generative replay and prevents the imbalance problem. Our approach is computationally efficient and scalable to large datasets. Experiments confirm that our approach achieves state-of-the-art results on CIFAR-100 and ImageNet, while requiring only a fraction of the storage needed for exemplar-based continual learning. Code available at \\url{https://github.com/xialeiliu/GFR-IL}.", "authors": ["Xialei Liu", "Chenshen Wu", "Mikel Menta", "Luis Herranz", "Bogdan Raducanu", "Andrew D. Bagdanov", "Shangling Jui", "Joost van de Weijer"], "categories": ["cs.CV", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2020-04-20", "url": "https://arxiv.org/abs/2004.09199", "pdf_url": "https://arxiv.org/pdf/2004.09199v1", "arxiv_id": "2004.09199", "doi": "10.1109/CVPRW50498.2020.00121", "citation_count": 187, "influential_citation_count": 12, "has_code": true, "code_url": "https://github.com/xialeiliu/GFR-IL}", "venue": null, "quality_score": 0.5685}
{"id": "574e3dd459a738206b62ce7e27a860e0dad8d70d63e63be7aa6bf0ccab1134e4", "sources": ["arxiv", "semantic_scholar"], "title": "Continual Learning for Anomaly Detection in Surveillance Videos", "abstract": "Anomaly detection in surveillance videos has been recently gaining attention. A challenging aspect of high-dimensional applications such as video surveillance is continual learning. While current state-of-the-art deep learning approaches perform well on existing public datasets, they fail to work in a continual learning framework due to computational and storage issues. Furthermore, online decision making is an important but mostly neglected factor in this domain. Motivated by these research gaps, we propose an online anomaly detection method for surveillance videos using transfer learning and continual learning, which in turn significantly reduces the training complexity and provides a mechanism for continually learning from recent data without suffering from catastrophic forgetting. Our proposed algorithm leverages the feature extraction power of neural network-based models for transfer learning, and the continual learning capability of statistical detection methods.", "authors": ["Keval Doshi", "Yasin Yilmaz"], "categories": ["cs.CV", "cs.LG", "eess.IV", "stat.ML"], "fields_of_study": ["Computer Science", "Engineering", "Mathematics"], "published_date": "2020-04-15", "url": "https://arxiv.org/abs/2004.07941", "pdf_url": "https://arxiv.org/pdf/2004.07941v1", "arxiv_id": "2004.07941", "doi": "10.1109/CVPRW50498.2020.00135", "citation_count": 138, "influential_citation_count": 12, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.557}
{"id": "9ee64cc2f4ba157093b78df564462f8f3e2e2b7939048df1015fd74db9dcd101", "sources": ["arxiv", "semantic_scholar"], "title": "A Demonstration of Issues with Value-Based Multiobjective Reinforcement Learning Under Stochastic State Transitions", "abstract": "We report a previously unidentified issue with model-free, value-based approaches to multiobjective reinforcement learning in the context of environments with stochastic state transitions. An example multiobjective Markov Decision Process (MOMDP) is used to demonstrate that under such conditions these approaches may be unable to discover the policy which maximises the Scalarised Expected Return, and in fact may converge to a Pareto-dominated solution. We discuss several alternative methods which may be more suitable for maximising SER in MOMDPs with stochastic transitions.", "authors": ["Peter Vamplew", "Cameron Foale", "Richard Dazeley"], "categories": ["cs.LG", "cs.MA", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-04-14", "url": "https://arxiv.org/abs/2004.06277", "pdf_url": "https://arxiv.org/pdf/2004.06277v1", "arxiv_id": "2004.06277", "doi": null, "citation_count": 3, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1505}
{"id": "f28c997677af5a7fc94b193bbf1bdfb7b0133303460c93824adfb2fa697d25ba", "sources": ["arxiv", "semantic_scholar"], "title": "A Learning Framework for n-bit Quantized Neural Networks toward FPGAs", "abstract": "The quantized neural network (QNN) is an efficient approach for network compression and can be widely used in the implementation of FPGAs. This paper proposes a novel learning framework for n-bit QNNs, whose weights are constrained to the power of two. To solve the gradient vanishing problem, we propose a reconstructed gradient function for QNNs in back-propagation algorithm that can directly get the real gradient rather than estimating an approximate gradient of the expected loss. We also propose a novel QNN structure named n-BQ-NN, which uses shift operation to replace the multiply operation and is more suitable for the inference on FPGAs. Furthermore, we also design a shift vector processing element (SVPE) array to replace all 16-bit multiplications with SHIFT operations in convolution operation on FPGAs. We also carry out comparable experiments to evaluate our framework. The experimental results show that the quantized models of ResNet, DenseNet and AlexNet through our learning framework can achieve almost the same accuracies with the original full-precision models. Moreover, when using our learning framework to train our n-BQ-NN from scratch, it can achieve state-of-the-art results compared with typical low-precision QNNs. Experiments on Xilinx ZCU102 platform show that our n-BQ-NN with our SVPE can execute 2.9 times faster than with the vector processing element (VPE) in inference. As the SHIFT operation in our SVPE array will not consume Digital Signal Processings (DSPs) resources on FPGAs, the experiments have shown that the use of SVPE array also reduces average energy consumption to 68.7% of the VPE array with 16-bit.", "authors": ["Jun Chen", "Liang Liu", "Yong Liu", "Xianfang Zeng"], "categories": ["cs.LG", "eess.SP", "stat.ML"], "fields_of_study": ["Computer Science", "Medicine", "Mathematics"], "published_date": "2020-04-06", "url": "https://arxiv.org/abs/2004.02396", "pdf_url": "https://arxiv.org/pdf/2004.02396v1", "arxiv_id": "2004.02396", "doi": "10.1109/TNNLS.2020.2980041", "citation_count": 34, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.386}
{"id": "88791ebfe16dbcdabc58e6ed7b299f3d7ab9fe55c15570fa18b52a8ef622ec93", "sources": ["arxiv", "semantic_scholar"], "title": "Learning Latent Causal Structures with a Redundant Input Neural Network", "abstract": "Most causal discovery algorithms find causal structure among a set of observed variables. Learning the causal structure among latent variables remains an important open problem, particularly when using high-dimensional data. In this paper, we address a problem for which it is known that inputs cause outputs, and these causal relationships are encoded by a causal network among a set of an unknown number of latent variables. We developed a deep learning model, which we call a redundant input neural network (RINN), with a modified architecture and a regularized objective function to find causal relationships between input, hidden, and output variables. More specifically, our model allows input variables to directly interact with all latent variables in a neural network to influence what information the latent variables should encode in order to generate the output variables accurately. In this setting, the direct connections between input and latent variables makes the latent variables partially interpretable; furthermore, the connectivity among the latent variables in the neural network serves to model their potential causal relationships to each other and to the output variables. A series of simulation experiments provide support that the RINN method can successfully recover latent causal structure between input and output variables.", "authors": ["Jonathan D. Young", "Bryan Andrews", "Gregory F. Cooper", "Xinghua Lu"], "categories": ["cs.LG", "cs.NE", "q-bio.MN", "stat.ML"], "fields_of_study": ["Computer Science", "Biology"], "published_date": "2020-03-29", "url": "https://arxiv.org/abs/2003.13135", "pdf_url": "https://arxiv.org/pdf/2003.13135v3", "arxiv_id": "2003.13135", "doi": null, "citation_count": 10, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2603}
{"id": "c171f562834e9f1a668e4542634f673016d577bb45a66d3d3cd608e8c4896e12", "sources": ["arxiv", "semantic_scholar"], "title": "Learning representations in Bayesian Confidence Propagation neural networks", "abstract": "Unsupervised learning of hierarchical representations has been one of the most vibrant research directions in deep learning during recent years. In this work we study biologically inspired unsupervised strategies in neural networks based on local Hebbian learning. We propose new mechanisms to extend the Bayesian Confidence Propagating Neural Network (BCPNN) architecture, and demonstrate their capability for unsupervised learning of salient hidden representations when tested on the MNIST dataset.", "authors": ["Naresh Balaji Ravichandran", "Anders Lansner", "Pawel Herman"], "categories": ["cs.LG", "cs.NE", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-03-27", "url": "https://arxiv.org/abs/2003.12415", "pdf_url": "https://arxiv.org/pdf/2003.12415v1", "arxiv_id": "2003.12415", "doi": "10.1109/IJCNN48605.2020.9207061", "citation_count": 15, "influential_citation_count": 3, "has_code": false, "code_url": null, "venue": "IEEE International Joint Conference on Neural Network", "quality_score": 0.301}
{"id": "74b4dfd686df4102faf6191bec566e72f2876327b4bb51f6f60b0031078424c1", "sources": ["arxiv", "semantic_scholar"], "title": "Overcoming Catastrophic Forgetting in Graph Neural Networks with Experience Replay", "abstract": "Graph Neural Networks (GNNs) have recently received significant research attention due to their superior performance on a variety of graph-related learning tasks. Most of the current works focus on either static or dynamic graph settings, addressing a single particular task, e.g., node/graph classification, link prediction. In this work, we investigate the question: can GNNs be applied to continuously learning a sequence of tasks? Towards that, we explore the Continual Graph Learning (CGL) paradigm and present the Experience Replay based framework ER-GNN for CGL to alleviate the catastrophic forgetting problem in existing GNNs. ER-GNN stores knowledge from previous tasks as experiences and replays them when learning new tasks to mitigate the catastrophic forgetting issue. We propose three experience node selection strategies: mean of feature, coverage maximization, and influence maximization, to guide the process of selecting experience nodes. Extensive experiments on three benchmark datasets demonstrate the effectiveness of our ER-GNN and shed light on the incremental graph (non-Euclidean) structure learning.", "authors": ["Fan Zhou", "Chengtai Cao"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-03-22", "url": "https://arxiv.org/abs/2003.09908", "pdf_url": "https://arxiv.org/pdf/2003.09908v2", "arxiv_id": "2003.09908", "doi": null, "citation_count": 10, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2603}
{"id": "cf67de2b3ae4a3981b97bfb3f01333895f8e1d9eb0beb442d2426f096638f533", "sources": ["arxiv", "semantic_scholar"], "title": "On Information Plane Analyses of Neural Network Classifiers -- A Review", "abstract": "We review the current literature concerned with information plane analyses of neural network classifiers. While the underlying information bottleneck theory and the claim that information-theoretic compression is causally linked to generalization are plausible, empirical evidence was found to be both supporting and conflicting. We review this evidence together with a detailed analysis of how the respective information quantities were estimated. Our survey suggests that compression visualized in information planes is not necessarily information-theoretic, but is rather often compatible with geometric compression of the latent representations. This insight gives the information plane a renewed justification. Aside from this, we shed light on the problem of estimating mutual information in deterministic neural networks and its consequences. Specifically, we argue that even in feed-forward neural networks the data processing inequality need not hold for estimates of mutual information. Similarly, while a fitting phase, in which the mutual information between the latent representation and the target increases, is necessary (but not sufficient) for good classification performance, depending on the specifics of mutual information estimation such a fitting phase need not be visible in the information plane.", "authors": ["Bernhard C. Geiger"], "categories": ["cs.LG", "cs.CV", "cs.IT", "stat.ML"], "fields_of_study": ["Medicine", "Computer Science", "Mathematics"], "published_date": "2020-03-21", "url": "https://arxiv.org/abs/2003.09671", "pdf_url": "https://arxiv.org/pdf/2003.09671v3", "arxiv_id": "2003.09671", "doi": "10.1109/TNNLS.2021.3089037", "citation_count": 63, "influential_citation_count": 8, "has_code": false, "code_url": null, "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.4771}
{"id": "c9dcad6ca127e63c3fe20d3a2c54fb651be751dfb7855616fe61da836db73e82", "sources": ["arxiv", "semantic_scholar"], "title": "Online Continual Learning on Sequences", "abstract": "Online continual learning (OCL) refers to the ability of a system to learn over time from a continuous stream of data without having to revisit previously encountered training samples. Learning continually in a single data pass is crucial for agents and robots operating in changing environments and required to acquire, fine-tune, and transfer increasingly complex representations from non-i.i.d. input distributions. Machine learning models that address OCL must alleviate \\textit{catastrophic forgetting} in which hidden representations are disrupted or completely overwritten when learning from streams of novel input. In this chapter, we summarize and discuss recent deep learning models that address OCL on sequential input through the use (and combination) of synaptic regularization, structural plasticity, and experience replay. Different implementations of replay have been proposed that alleviate catastrophic forgetting in connectionists architectures via the re-occurrence of (latent representations of) input sequences and that functionally resemble mechanisms of hippocampal replay in the mammalian brain. Empirical evidence shows that architectures endowed with experience replay typically outperform architectures without in (online) incremental learning tasks.", "authors": ["German I. Parisi", "Vincenzo Lomonaco"], "categories": ["cs.LG", "cs.CV", "cs.NE"], "fields_of_study": ["Computer Science"], "published_date": "2020-03-20", "url": "https://arxiv.org/abs/2003.09114", "pdf_url": "https://arxiv.org/pdf/2003.09114v1", "arxiv_id": "2003.09114", "doi": "10.1007/978-3-030-43883-8_8", "citation_count": 35, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.3891}
{"id": "43f22790170c53370082f83ba1a64a1bde98df84c678d2e697084dceb5e0dbe1", "sources": ["arxiv", "semantic_scholar"], "title": "Lifelong Learning with Searchable Extension Units", "abstract": "Lifelong learning remains an open problem. One of its main difficulties is catastrophic forgetting. Many dynamic expansion approaches have been proposed to address this problem, but they all use homogeneous models of predefined structure for all tasks. The common original model and expansion structures ignore the requirement of different model structures on different tasks, which leads to a less compact model for multiple tasks and causes the model size to increase rapidly as the number of tasks increases. Moreover, they can not perform best on all tasks. To solve those problems, in this paper, we propose a new lifelong learning framework named Searchable Extension Units (SEU) by introducing Neural Architecture Search into lifelong learning, which breaks down the need for a predefined original model and searches for specific extension units for different tasks, without compromising the performance of the model on different tasks. Our approach can obtain a much more compact model without catastrophic forgetting. The experimental results on the PMNIST, the split CIFAR10 dataset, the split CIFAR100 dataset, and the Mixture dataset empirically prove that our method can achieve higher accuracy with much smaller model, whose size is about 25-33 percentage of that of the state-of-the-art methods.", "authors": ["Wenjin Wang", "Yunqing Hu", "Yin Zhang"], "categories": ["cs.LG", "cs.CV", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-03-19", "url": "https://arxiv.org/abs/2003.08559", "pdf_url": "https://arxiv.org/pdf/2003.08559v1", "arxiv_id": "2003.08559", "doi": null, "citation_count": 3, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1505}
{"id": "07d63eeb0363545cd824b44629aec9185d4274322b98658021054ea50912cda6", "sources": ["arxiv", "semantic_scholar"], "title": "XtarNet: Learning to Extract Task-Adaptive Representation for Incremental Few-Shot Learning", "abstract": "Learning novel concepts while preserving prior knowledge is a long-standing challenge in machine learning. The challenge gets greater when a novel task is given with only a few labeled examples, a problem known as incremental few-shot learning. We propose XtarNet, which learns to extract task-adaptive representation (TAR) for facilitating incremental few-shot learning. The method utilizes a backbone network pretrained on a set of base categories while also employing additional modules that are meta-trained across episodes. Given a new task, the novel feature extracted from the meta-trained modules is mixed with the base feature obtained from the pretrained model. The process of combining two different features provides TAR and is also controlled by meta-trained modules. The TAR contains effective information for classifying both novel and base categories. The base and novel classifiers quickly adapt to a given task by utilizing the TAR. Experiments on standard image datasets indicate that XtarNet achieves state-of-the-art incremental few-shot learning performance. The concept of TAR can also be used in conjunction with existing incremental few-shot learning methods; extensive simulation results in fact show that applying TAR enhances the known methods significantly.", "authors": ["Sung Whan Yoon", "Do-Yeon Kim", "Jun Seo", "Jaekyun Moon"], "categories": ["cs.LG", "cs.AI", "cs.NE"], "fields_of_study": ["Computer Science"], "published_date": "2020-03-19", "url": "https://arxiv.org/abs/2003.08561", "pdf_url": "https://arxiv.org/pdf/2003.08561v2", "arxiv_id": "2003.08561", "doi": null, "citation_count": 49, "influential_citation_count": 8, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.4771}
{"id": "c65e557bc9c4065482fdbb7dcf5c9cc696b2a5543c99c171a6492f96adc2b6b3", "sources": ["arxiv", "semantic_scholar"], "title": "Tensor Graph Convolutional Networks for Multi-relational and Robust Learning", "abstract": "The era of \"data deluge\" has sparked renewed interest in graph-based learning methods and their widespread applications ranging from sociology and biology to transportation and communications. In this context of graph-aware methods, the present paper introduces a tensor-graph convolutional network (TGCN) for scalable semi-supervised learning (SSL) from data associated with a collection of graphs, that are represented by a tensor. Key aspects of the novel TGCN architecture are the dynamic adaptation to different relations in the tensor graph via learnable weights, and the consideration of graph-based regularizers to promote smoothness and alleviate over-parameterization. The ultimate goal is to design a powerful learning architecture able to: discover complex and highly nonlinear data associations, combine (and select) multiple types of relations, scale gracefully with the graph size, and remain robust to perturbations on the graph edges. The proposed architecture is relevant not only in applications where the nodes are naturally involved in different relations (e.g., a multi-relational graph capturing family, friendship and work relations in a social network), but also in robust learning setups where the graph entails a certain level of uncertainty, and the different tensor slabs correspond to different versions (realizations) of the nominal graph. Numerical tests showcase that the proposed architecture achieves markedly improved performance relative to standard GCNs, copes with state-of-the-art adversarial attacks, and leads to remarkable SSL performance over protein-to-protein interaction networks.", "authors": ["Vassilis N. Ioannidis", "Antonio G. Marques", "Georgios B. Giannakis"], "categories": ["cs.LG", "eess.SP", "stat.ML"], "fields_of_study": ["Computer Science", "Engineering", "Mathematics", "Sociology"], "published_date": "2020-03-15", "url": "https://arxiv.org/abs/2003.07729", "pdf_url": "https://arxiv.org/pdf/2003.07729v1", "arxiv_id": "2003.07729", "doi": "10.1109/TSP.2020.3028495", "citation_count": 30, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "IEEE Transactions on Signal Processing", "quality_score": 0.3728}
{"id": "7d6c52929418f2a6a436395802f4e22711b6ddddc87d3a19115ffd8cc27896ab", "sources": ["arxiv", "semantic_scholar"], "title": "Triple Memory Networks: a Brain-Inspired Method for Continual Learning", "abstract": "Continual acquisition of novel experience without interfering previously learned knowledge, i.e. continual learning, is critical for artificial neural networks, but limited by catastrophic forgetting. A neural network adjusts its parameters when learning a new task, but then fails to conduct the old tasks well. By contrast, the brain has a powerful ability to continually learn new experience without catastrophic interference. The underlying neural mechanisms possibly attribute to the interplay of hippocampus-dependent memory system and neocortex-dependent memory system, mediated by prefrontal cortex. Specifically, the two memory systems develop specialized mechanisms to consolidate information as more specific forms and more generalized forms, respectively, and complement the two forms of information in the interplay. Inspired by such brain strategy, we propose a novel approach named triple memory networks (TMNs) for continual learning. TMNs model the interplay of hippocampus, prefrontal cortex and sensory cortex (a neocortex region) as a triple-network architecture of generative adversarial networks (GAN). The input information is encoded as specific representation of the data distributions in a generator, or generalized knowledge of solving tasks in a discriminator and a classifier, with implementing appropriate brain-inspired algorithms to alleviate catastrophic forgetting in each module. Particularly, the generator replays generated data of the learned tasks to the discriminator and the classifier, both of which are implemented with a weight consolidation regularizer to complement the lost information in generation process. TMNs achieve new state-of-the-art performance on a variety of class-incremental learning benchmarks on MNIST, SVHN, CIFAR-10 and ImageNet-50, comparing with strong baseline methods.", "authors": ["Liyuan Wang", "Bo Lei", "Qian Li", "Hang Su", "Jun Zhu", "Yi Zhong"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics", "Medicine"], "published_date": "2020-03-06", "url": "https://arxiv.org/abs/2003.03143", "pdf_url": "https://arxiv.org/pdf/2003.03143v1", "arxiv_id": "2003.03143", "doi": "10.1109/TNNLS.2021.3111019", "citation_count": 64, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.4532}
{"id": "164925faba84fe7828b2a03895b8e46b87aef2a585608b3de5ce54bab261b1cf", "sources": ["arxiv", "semantic_scholar"], "title": "Adaptive Propagation Graph Convolutional Network", "abstract": "Graph convolutional networks (GCNs) are a family of neural network models that perform inference on graph data by interleaving vertex-wise operations and message-passing exchanges across nodes. Concerning the latter, two key questions arise: (i) how to design a differentiable exchange protocol (e.g., a 1-hop Laplacian smoothing in the original GCN), and (ii) how to characterize the trade-off in complexity with respect to the local updates. In this paper, we show that state-of-the-art results can be achieved by adapting the number of communication steps independently at every node. In particular, we endow each node with a halting unit (inspired by Graves' adaptive computation time) that after every exchange decides whether to continue communicating or not. We show that the proposed adaptive propagation GCN (AP-GCN) achieves superior or similar results to the best proposed models so far on a number of benchmarks, while requiring a small overhead in terms of additional parameters. We also investigate a regularization term to enforce an explicit trade-off between communication and accuracy. The code for the AP-GCN experiments is released as an open-source library.", "authors": ["Indro Spinelli", "Simone Scardapane", "Aurelio Uncini"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Medicine", "Mathematics"], "published_date": "2020-02-24", "url": "https://arxiv.org/abs/2002.10306", "pdf_url": "https://arxiv.org/pdf/2002.10306v3", "arxiv_id": "2002.10306", "doi": "10.1109/TNNLS.2020.3025110", "citation_count": 95, "influential_citation_count": 6, "has_code": true, "code_url": null, "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.4956}
{"id": "8350530b1f4377ba67d9464a10ae3a3fb6a7953ce48c43b478ac724159b0e3f0", "sources": ["arxiv", "semantic_scholar"], "title": "Learning to Continually Learn", "abstract": "Continual lifelong learning requires an agent or model to learn many sequentially ordered tasks, building on previous knowledge without catastrophically forgetting it. Much work has gone towards preventing the default tendency of machine learning models to catastrophically forget, yet virtually all such work involves manually-designed solutions to the problem. We instead advocate meta-learning a solution to catastrophic forgetting, allowing AI to learn to continually learn. Inspired by neuromodulatory processes in the brain, we propose A Neuromodulated Meta-Learning Algorithm (ANML). It differentiates through a sequential learning process to meta-learn an activation-gating function that enables context-dependent selective activation within a deep neural network. Specifically, a neuromodulatory (NM) neural network gates the forward pass of another (otherwise normal) neural network called the prediction learning network (PLN). The NM network also thus indirectly controls selective plasticity (i.e. the backward pass of) the PLN. ANML enables continual learning without catastrophic forgetting at scale: it produces state-of-the-art continual learning performance, sequentially learning as many as 600 classes (over 9,000 SGD updates).", "authors": ["Shawn Beaulieu", "Lapo Frati", "Thomas Miconi", "Joel Lehman", "Kenneth O. Stanley", "Jeff Clune", "Nick Cheney"], "categories": ["cs.LG", "cs.CV", "cs.NE", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-02-21", "url": "https://arxiv.org/abs/2002.09571", "pdf_url": "https://arxiv.org/pdf/2002.09571v2", "arxiv_id": "2002.09571", "doi": "10.3233/FAIA200193", "citation_count": 166, "influential_citation_count": 27, "has_code": false, "code_url": null, "venue": "European Conference on Artificial Intelligence", "quality_score": 0.7236}
{"id": "ef1aeefaf217d7dca13fc584f05a48e2f1e922a0887a6f3de05a14743fd25167", "sources": ["arxiv", "semantic_scholar"], "title": "BatchEnsemble: An Alternative Approach to Efficient Ensemble and Lifelong Learning", "abstract": "Ensembles, where multiple neural networks are trained individually and their predictions are averaged, have been shown to be widely successful for improving both the accuracy and predictive uncertainty of single neural networks. However, an ensemble's cost for both training and testing increases linearly with the number of networks, which quickly becomes untenable. In this paper, we propose BatchEnsemble, an ensemble method whose computational and memory costs are significantly lower than typical ensembles. BatchEnsemble achieves this by defining each weight matrix to be the Hadamard product of a shared weight among all ensemble members and a rank-one matrix per member. Unlike ensembles, BatchEnsemble is not only parallelizable across devices, where one device trains one member, but also parallelizable within a device, where multiple ensemble members are updated simultaneously for a given mini-batch. Across CIFAR-10, CIFAR-100, WMT14 EN-DE/EN-FR translation, and out-of-distribution tasks, BatchEnsemble yields competitive accuracy and uncertainties as typical ensembles; the speedup at test time is 3X and memory reduction is 3X at an ensemble of size 4. We also apply BatchEnsemble to lifelong learning, where on Split-CIFAR-100, BatchEnsemble yields comparable performance to progressive neural networks while having a much lower computational and memory costs. We further show that BatchEnsemble can easily scale up to lifelong learning on Split-ImageNet which involves 100 sequential learning tasks.", "authors": ["Yeming Wen", "Dustin Tran", "Jimmy Ba"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-02-17", "url": "https://arxiv.org/abs/2002.06715", "pdf_url": "https://arxiv.org/pdf/2002.06715v2", "arxiv_id": "2002.06715", "doi": null, "citation_count": 581, "influential_citation_count": 70, "has_code": false, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.9256}
{"id": "cab8086b3b36528dbf1120f251f3ea0518ca1122b393c4c2dc6d7009bfd7ec74", "sources": ["arxiv", "semantic_scholar"], "title": "RNA Secondary Structure Prediction By Learning Unrolled Algorithms", "abstract": "In this paper, we propose an end-to-end deep learning model, called E2Efold, for RNA secondary structure prediction which can effectively take into account the inherent constraints in the problem. The key idea of E2Efold is to directly predict the RNA base-pairing matrix, and use an unrolled algorithm for constrained programming as the template for deep architectures to enforce constraints. With comprehensive experiments on benchmark datasets, we demonstrate the superior performance of E2Efold: it predicts significantly better structures compared to previous SOTA (especially for pseudoknotted structures), while being as efficient as the fastest algorithms in terms of inference time.", "authors": ["Xinshi Chen", "Yu Li", "Ramzan Umarov", "Xin Gao", "Le Song"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-02-13", "url": "https://arxiv.org/abs/2002.05810", "pdf_url": "https://arxiv.org/pdf/2002.05810v1", "arxiv_id": "2002.05810", "doi": null, "citation_count": 135, "influential_citation_count": 24, "has_code": false, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.699}
{"id": "2d65fe9c33c7f06134966e59432fd02bd3e873bd218778810cb87dd52a544080", "sources": ["arxiv", "semantic_scholar"], "title": "Learning Flat Latent Manifolds with VAEs", "abstract": "Measuring the similarity between data points often requires domain knowledge, which can in parts be compensated by relying on unsupervised methods such as latent-variable models, where similarity/distance is estimated in a more compact latent space. Prevalent is the use of the Euclidean metric, which has the drawback of ignoring information about similarity of data stored in the decoder, as captured by the framework of Riemannian geometry. We propose an extension to the framework of variational auto-encoders allows learning flat latent manifolds, where the Euclidean metric is a proxy for the similarity between data points. This is achieved by defining the latent space as a Riemannian manifold and by regularising the metric tensor to be a scaled identity matrix. Additionally, we replace the compact prior typically used in variational auto-encoders with a recently presented, more expressive hierarchical one---and formulate the learning problem as a constrained optimisation problem. We evaluate our method on a range of data-sets, including a video-tracking benchmark, where the performance of our unsupervised approach nears that of state-of-the-art supervised approaches, while retaining the computational efficiency of straight-line-based approaches.", "authors": ["Nutan Chen", "Alexej Klushyn", "Francesco Ferroni", "Justin Bayer", "Patrick van der Smagt"], "categories": ["stat.ML", "cs.LG"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-02-12", "url": "https://arxiv.org/abs/2002.04881", "pdf_url": "https://arxiv.org/pdf/2002.04881v3", "arxiv_id": "2002.04881", "doi": null, "citation_count": 52, "influential_citation_count": 4, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.4311}
{"id": "f992a8322c3d69fdbd8c2a571f3b883b96780517c5b4efa620ad6a736f3c9692", "sources": ["arxiv", "semantic_scholar"], "title": "Spiking Inception Module for Multi-layer Unsupervised Spiking Neural Networks", "abstract": "Spiking Neural Network (SNN), as a brain-inspired approach, is attracting attention due to its potential to produce ultra-high-energy-efficient hardware. Competitive learning based on Spike-Timing-Dependent Plasticity (STDP) is a popular method to train an unsupervised SNN. However, previous unsupervised SNNs trained through this method are limited to a shallow network with only one learnable layer and cannot achieve satisfactory results when compared with multi-layer SNNs. In this paper, we eased this limitation by: 1)We proposed a Spiking Inception (Sp-Inception) module, inspired by the Inception module in the Artificial Neural Network (ANN) literature. This module is trained through STDP-based competitive learning and outperforms the baseline modules on learning capability, learning efficiency, and robustness. 2)We proposed a Pooling-Reshape-Activate (PRA) layer to make the Sp-Inception module stackable. 3)We stacked multiple Sp-Inception modules to construct multi-layer SNNs. Our algorithm outperforms the baseline algorithms on the hand-written digit classification task, and reaches state-of-the-art results on the MNIST dataset among the existing unsupervised SNNs.", "authors": ["Mingyuan Meng", "Xingyu Yang", "Shanlin Xiao", "Zhiyi Yu"], "categories": ["cs.NE", "cs.LG", "q-bio.NC"], "fields_of_study": ["Computer Science"], "published_date": "2020-01-29", "url": "https://arxiv.org/abs/2001.10696", "pdf_url": "https://arxiv.org/pdf/2001.10696v5", "arxiv_id": "2001.10696", "doi": "10.1109/IJCNN48605.2020.9207161", "citation_count": 14, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "IEEE International Joint Conference on Neural Network", "quality_score": 0.294}
{"id": "92b7650f182ed41cce59668c27df57556758acd8390aa09d5321e7305cac98f7", "sources": ["arxiv", "semantic_scholar"], "title": "Supervised Learning for Non-Sequential Data: A Canonical Polyadic Decomposition Approach", "abstract": "Efficient modelling of feature interactions underpins supervised learning for non-sequential tasks, characterized by a lack of inherent ordering of features (variables). The brute force approach of learning a parameter for each interaction of every order comes at an exponential computational and memory cost (Curse of Dimensionality). To alleviate this issue, it has been proposed to implicitly represent the model parameters as a tensor, the order of which is equal to the number of features; for efficiency, it can be further factorized into a compact Tensor Train (TT) format. However, both TT and other Tensor Networks (TNs), such as Tensor Ring and Hierarchical Tucker, are sensitive to the ordering of their indices (and hence to the features). To establish the desired invariance to feature ordering, we propose to represent the weight tensor through the Canonical Polyadic (CP) Decomposition (CPD), and introduce the associated inference and learning algorithms, including suitable regularization and initialization schemes. It is demonstrated that the proposed CP-based predictor significantly outperforms other TN-based predictors on sparse data while exhibiting comparable performance on dense non-sequential tasks. Furthermore, for enhanced expressiveness, we generalize the framework to allow feature mapping to arbitrarily high-dimensional feature vectors. In conjunction with feature vector normalization, this is shown to yield dramatic improvements in performance for dense non-sequential tasks, matching models such as fully-connected neural networks.", "authors": ["Alexandros Haliassos", "Kriton Konstantinidis", "Danilo P. Mandic"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-01-27", "url": "https://arxiv.org/abs/2001.10109", "pdf_url": "https://arxiv.org/pdf/2001.10109v3", "arxiv_id": "2001.10109", "doi": null, "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.0753}
{"id": "01aa1409c4ce7de965c3049407ffb0f67fcaf527f5247511f9a8ffbd78f268b8", "sources": ["arxiv", "semantic_scholar"], "title": "Uncertainty-based Modulation for Lifelong Learning", "abstract": "The creation of machine learning algorithms for intelligent agents capable of continuous, lifelong learning is a critical objective for algorithms being deployed on real-life systems in dynamic environments. Here we present an algorithm inspired by neuromodulatory mechanisms in the human brain that integrates and expands upon Stephen Grossbergś ground-breaking Adaptive Resonance Theory proposals. Specifically, it builds on the concept of uncertainty, and employs a series of neuromodulatory mechanisms to enable continuous learning, including self-supervised and one-shot learning. Algorithm components were evaluated in a series of benchmark experiments that demonstrate stable learning without catastrophic forgetting. We also demonstrate the critical role of developing these systems in a closed-loop manner where the environment and the agentś behaviors constrain and guide the learning process. To this end, we integrated the algorithm into an embodied simulated drone agent. The experiments show that the algorithm is capable of continuous learning of new tasks and under changed conditions with high classification accuracy (greater than 94 percent) in a virtual environment, without catastrophic forgetting. The algorithm accepts high dimensional inputs from any state-of-the-art detection and feature extraction algorithms, making it a flexible addition to existing systems. We also describe future development efforts focused on imbuing the algorithm with mechanisms to seek out new knowledge as well as employ a broader range of neuromodulatory processes.", "authors": ["Andrew Brna", "Ryan Brown", "Patrick Connolly", "Stephen Simons", "Renee Shimizu", "Mario Aguilar-Simon"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics", "Medicine"], "published_date": "2020-01-27", "url": "https://arxiv.org/abs/2001.09822", "pdf_url": "https://arxiv.org/pdf/2001.09822v1", "arxiv_id": "2001.09822", "doi": "10.1016/j.neunet.2019.09.011", "citation_count": 22, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "Neural Networks", "quality_score": 0.3404}
{"id": "702863f9f016bd765eb7659448ea38665dddcdfaad861b47cf270cdeb5db3f8a", "sources": ["arxiv", "semantic_scholar"], "title": "Dissecting Catastrophic Forgetting in Continual Learning by Deep Visualization", "abstract": "Interpreting the behaviors of Deep Neural Networks (usually considered as a black box) is critical especially when they are now being widely adopted over diverse aspects of human life. Taking the advancements from Explainable Artificial Intelligent, this paper proposes a novel technique called Auto DeepVis to dissect catastrophic forgetting in continual learning. A new method to deal with catastrophic forgetting named critical freezing is also introduced upon investigating the dilemma by Auto DeepVis. Experiments on a captioning model meticulously present how catastrophic forgetting happens, particularly showing which components are forgetting or changing. The effectiveness of our technique is then assessed; and more precisely, critical freezing claims the best performance on both previous and coming tasks over baselines, proving the capability of the investigation. Our techniques could not only be supplementary to existing solutions for completely eradicating catastrophic forgetting for life-long learning but also explainable.", "authors": ["Giang Nguyen", "Shuan Chen", "Thao Do", "Tae Joon Jun", "Ho-Jin Choi", "Daeyoung Kim"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-01-06", "url": "https://arxiv.org/abs/2001.01578", "pdf_url": "https://arxiv.org/pdf/2001.01578v2", "arxiv_id": "2001.01578", "doi": null, "citation_count": 13, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2865}
{"id": "0ef7b74750afbeaed305052d1a51c7e0e0ed2c1bf7a04c7e3a169d31dae843d9", "sources": ["arxiv", "semantic_scholar"], "title": "A Neural Dirichlet Process Mixture Model for Task-Free Continual Learning", "abstract": "Despite the growing interest in continual learning, most of its contemporary works have been studied in a rather restricted setting where tasks are clearly distinguishable, and task boundaries are known during training. However, if our goal is to develop an algorithm that learns as humans do, this setting is far from realistic, and it is essential to develop a methodology that works in a task-free manner. Meanwhile, among several branches of continual learning, expansion-based methods have the advantage of eliminating catastrophic forgetting by allocating new resources to learn new data. In this work, we propose an expansion-based approach for task-free continual learning. Our model, named Continual Neural Dirichlet Process Mixture (CN-DPM), consists of a set of neural network experts that are in charge of a subset of the data. CN-DPM expands the number of experts in a principled way under the Bayesian nonparametric framework. With extensive experiments, we show that our model successfully performs task-free continual learning for both discriminative and generative tasks such as image classification and image generation.", "authors": ["Soochan Lee", "Junsoo Ha", "Dongsu Zhang", "Gunhee Kim"], "categories": ["cs.LG", "cs.NE", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2020-01-03", "url": "https://arxiv.org/abs/2001.00689", "pdf_url": "https://arxiv.org/pdf/2001.00689v2", "arxiv_id": "2001.00689", "doi": null, "citation_count": 240, "influential_citation_count": 32, "has_code": false, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.7593}
{"id": "aaca9f8e3d9fb7a8de36ea282101f80d3eca7de78cc627763b4082c1f585d071", "sources": ["arxiv", "semantic_scholar"], "title": "Overcoming Long-term Catastrophic Forgetting through Adversarial Neural Pruning and Synaptic Consolidation", "abstract": "Artificial neural networks face the well-known problem of catastrophic forgetting. What's worse, the degradation of previously learned skills becomes more severe as the task sequence increases, known as the long-term catastrophic forgetting. It is due to two facts: first, as the model learns more tasks, the intersection of the low-error parameter subspace satisfying for these tasks becomes smaller or even does not exist; second, when the model learns a new task, the cumulative error keeps increasing as the model tries to protect the parameter configuration of previous tasks from interference. Inspired by the memory consolidation mechanism in mammalian brains with synaptic plasticity, we propose a confrontation mechanism in which Adversarial Neural Pruning and synaptic Consolidation (ANPyC) is used to overcome the long-term catastrophic forgetting issue. The neural pruning acts as long-term depression to prune task-irrelevant parameters, while the novel synaptic consolidation acts as long-term potentiation to strengthen task-relevant parameters. During the training, this confrontation achieves a balance in that only crucial parameters remain, and non-significant parameters are freed to learn subsequent tasks. ANPyC avoids forgetting important information and makes the model efficient to learn a large number of tasks. Specifically, the neural pruning iteratively relaxes the current task's parameter conditions to expand the common parameter subspace of the task; the synaptic consolidation strategy, which consists of a structure-aware parameter-importance measurement and an element-wise parameter updating strategy, decreases the cumulative error when learning new tasks. The full source code is available at https://github.com/GeoX-Lab/ANPyC.", "authors": ["Jian Peng", "Bo Tang", "Hao Jiang", "Zhuo Li", "Yinjie Lei", "Tao Lin", "Haifeng Li"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics", "Medicine"], "published_date": "2019-12-19", "url": "https://arxiv.org/abs/1912.09091", "pdf_url": "https://arxiv.org/pdf/1912.09091v3", "arxiv_id": "1912.09091", "doi": "10.1109/TNNLS.2021.3056201", "citation_count": 47, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/GeoX-Lab/ANPyC", "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.4203}
{"id": "00243c2e1a9f27fa76fcac4e3d7ab717eb51bf6774a36b32b026d1165d89994f", "sources": ["arxiv", "semantic_scholar"], "title": "Deep Iterative and Adaptive Learning for Graph Neural Networks", "abstract": "In this paper, we propose an end-to-end graph learning framework, namely Deep Iterative and Adaptive Learning for Graph Neural Networks (DIAL-GNN), for jointly learning the graph structure and graph embeddings simultaneously. We first cast the graph structure learning problem as a similarity metric learning problem and leverage an adapted graph regularization for controlling smoothness, connectivity and sparsity of the generated graph. We further propose a novel iterative method for searching for a hidden graph structure that augments the initial graph structure. Our iterative method dynamically stops when the learned graph structure approaches close enough to the optimal graph. Our extensive experiments demonstrate that the proposed DIAL-GNN model can consistently outperform or match state-of-the-art baselines in terms of both downstream task performance and computational time. The proposed approach can cope with both transductive learning and inductive learning.", "authors": ["Yu Chen", "Lingfei Wu", "Mohammed J. Zaki"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-12-17", "url": "https://arxiv.org/abs/1912.07832", "pdf_url": "https://arxiv.org/pdf/1912.07832v1", "arxiv_id": "1912.07832", "doi": null, "citation_count": 52, "influential_citation_count": 3, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4311}
{"id": "9dbabfe24d8ac72581b780533006a4be39772977a5906f8e0d0bc164a824f6fb", "sources": ["arxiv", "semantic_scholar"], "title": "Neural Networks as Geometric Chaotic Maps", "abstract": "The use of artificial neural networks as models of chaotic dynamics has been rapidly expanding. Still, a theoretical understanding of how neural networks learn chaos is lacking. Here, we employ a geometric perspective to show that neural networks can efficiently model chaotic dynamics by becoming structurally chaotic themselves. We first confirm neural network's efficiency in emulating chaos by showing that a parsimonious neural network trained only on few data points can reconstruct strange attractors, extrapolate outside training data boundaries, and accurately predict local divergence rates. We then posit that the trained network's map comprises sequential geometric stretching, rotation, and compression operations. These geometric operations indicate topological mixing and chaos, explaining why neural networks are naturally suitable to emulate chaotic dynamics.", "authors": ["Ziwei Li", "Sai Ravela"], "categories": ["cs.LG", "math.DS", "nlin.CD", "stat.ML"], "fields_of_study": ["Medicine", "Computer Science", "Mathematics", "Physics"], "published_date": "2019-12-11", "url": "https://arxiv.org/abs/1912.05081", "pdf_url": "https://arxiv.org/pdf/1912.05081v4", "arxiv_id": "1912.05081", "doi": "10.1109/TNNLS.2021.3087497", "citation_count": 11, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.2698}
{"id": "57ae3b30ac820499114cbebf025257f4517b16aff15d6614ad10396316e63da4", "sources": ["arxiv", "semantic_scholar"], "title": "Reducing Catastrophic Forgetting in Modular Neural Networks by Dynamic Information Balancing", "abstract": "Lifelong learning is a very important step toward realizing robust autonomous artificial agents. Neural networks are the main engine of deep learning, which is the current state-of-the-art technique in formulating adaptive artificial intelligent systems. However, neural networks suffer from catastrophic forgetting when stressed with the challenge of continual learning. We investigate how to exploit modular topology in neural networks in order to dynamically balance the information load between different modules by routing inputs based on the information content in each module so that information interference is minimized. Our dynamic information balancing (DIB) technique adapts a reinforcement learning technique to guide the routing of different inputs based on a reward signal derived from a measure of the information load in each module. Our empirical results show that DIB combined with elastic weight consolidation (EWC) regularization outperforms models with similar capacity and EWC regularization across different task formulations and datasets.", "authors": ["Mohammed Amer", "Tomás Maul"], "categories": ["cs.LG", "cs.NE", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-12-10", "url": "https://arxiv.org/abs/1912.04508", "pdf_url": "https://arxiv.org/pdf/1912.04508v1", "arxiv_id": "1912.04508", "doi": null, "citation_count": 7, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2258}
{"id": "d52f70e591189c611fc58e45a3e15c8e13122735bd27e10c51fb9d682d461693", "sources": ["arxiv", "semantic_scholar"], "title": "ChainerRL: A Deep Reinforcement Learning Library", "abstract": "In this paper, we introduce ChainerRL, an open-source deep reinforcement learning (DRL) library built using Python and the Chainer deep learning framework. ChainerRL implements a comprehensive set of DRL algorithms and techniques drawn from state-of-the-art research in the field. To foster reproducible research, and for instructional purposes, ChainerRL provides scripts that closely replicate the original papers' experimental settings and reproduce published benchmark results for several algorithms. Lastly, ChainerRL offers a visualization tool that enables the qualitative inspection of trained agents. The ChainerRL source code can be found on GitHub: https://github.com/chainer/chainerrl.", "authors": ["Yasuhiro Fujita", "Prabhat Nagarajan", "Toshiki Kataoka", "Takahiro Ishikawa"], "categories": ["cs.LG", "cs.AI", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-12-09", "url": "https://arxiv.org/abs/1912.03905", "pdf_url": "https://arxiv.org/pdf/1912.03905v2", "arxiv_id": "1912.03905", "doi": null, "citation_count": 144, "influential_citation_count": 8, "has_code": true, "code_url": "https://github.com/chainer/chainerrl", "venue": "Journal of machine learning research", "quality_score": 0.5403}
{"id": "5accab88f48db8bcc07c4cc107086b85ad215f2696221b0e51deb917c69e879a", "sources": ["arxiv", "semantic_scholar"], "title": "Quadratic Q-network for Learning Continuous Control for Autonomous Vehicles", "abstract": "Reinforcement Learning algorithms have recently been proposed to learn time-sequential control policies in the field of autonomous driving. Direct applications of Reinforcement Learning algorithms with discrete action space will yield unsatisfactory results at the operational level of driving where continuous control actions are actually required. In addition, the design of neural networks often fails to incorporate the domain knowledge of the targeting problem such as the classical control theories in our case. In this paper, we propose a hybrid model by combining Q-learning and classic PID (Proportion Integration Differentiation) controller for handling continuous vehicle control problems under dynamic driving environment. Particularly, instead of using a big neural network as Q-function approximation, we design a Quadratic Q-function over actions with multiple simple neural networks for finding optimal values within a continuous space. We also build an action network based on the domain knowledge of the control mechanism of a PID controller to guide the agent to explore optimal actions more efficiently.We test our proposed approach in simulation under two common but challenging driving situations, the lane change scenario and ramp merge scenario. Results show that the autonomous vehicle agent can successfully learn a smooth and efficient driving behavior in both situations.", "authors": ["Pin Wang", "Hanhan Li", "Ching-Yao Chan"], "categories": ["cs.LG", "cs.AI", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-11-29", "url": "https://arxiv.org/abs/1912.00074", "pdf_url": "https://arxiv.org/pdf/1912.00074v1", "arxiv_id": "1912.00074", "doi": null, "citation_count": 13, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2865}
{"id": "ba14faf4ceaa673f5556a1ed650386345ce87a65ca06cfd033e19e910c78c197", "sources": ["arxiv", "semantic_scholar"], "title": "Memory-Efficient Episodic Control Reinforcement Learning with Dynamic Online k-means", "abstract": "Recently, neuro-inspired episodic control (EC) methods have been developed to overcome the data-inefficiency of standard deep reinforcement learning approaches. Using non-/semi-parametric models to estimate the value function, they learn rapidly, retrieving cached values from similar past states. In realistic scenarios, with limited resources and noisy data, maintaining meaningful representations in memory is essential to speed up the learning and avoid catastrophic forgetting. Unfortunately, EC methods have a large space and time complexity. We investigate different solutions to these problems based on prioritising and ranking stored states, as well as online clustering techniques. We also propose a new dynamic online k-means algorithm that is both computationally-efficient and yields significantly better performance at smaller memory sizes; we validate this approach on classic reinforcement learning environments and Atari games.", "authors": ["Andrea Agostinelli", "Kai Arulkumaran", "Marta Sarrico", "Pierre Richemond", "Anil Anthony Bharath"], "categories": ["cs.LG", "cs.NE", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-11-21", "url": "https://arxiv.org/abs/1911.09560", "pdf_url": "https://arxiv.org/pdf/1911.09560v1", "arxiv_id": "1911.09560", "doi": null, "citation_count": 5, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1945}
{"id": "22c8c783fd13d06794e361c6d7da6402b62f60e631d19657e3a54f74e9cda460", "sources": ["arxiv", "semantic_scholar"], "title": "A Recurrent Probabilistic Neural Network with Dimensionality Reduction Based on Time-series Discriminant Component Analysis", "abstract": "This paper proposes a probabilistic neural network developed on the basis of time-series discriminant component analysis (TSDCA) that can be used to classify high-dimensional time-series patterns. TSDCA involves the compression of high-dimensional time series into a lower-dimensional space using a set of orthogonal transformations and the calculation of posterior probabilities based on a continuous-density hidden Markov model with a Gaussian mixture model expressed in the reduced-dimensional space. The analysis can be incorporated into a neural network, which is named a time-series discriminant component network (TSDCN), so that parameters of dimensionality reduction and classification can be obtained simultaneously as network coefficients according to a backpropagation through time-based learning algorithm with the Lagrange multiplier method. The TSDCN is considered to enable high-accuracy classification of high-dimensional time-series patterns and to reduce the computation time taken for network training. The validity of the TSDCN is demonstrated for high-dimensional artificial data and EEG signals in the experiments conducted during the study.", "authors": ["Hideaki Hayashi", "Taro Shibanoki", "Keisuke Shima", "Yuichi Kurita", "Toshio Tsuji"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Medicine", "Computer Science", "Mathematics"], "published_date": "2019-11-14", "url": "https://arxiv.org/abs/1911.06009", "pdf_url": "https://arxiv.org/pdf/1911.06009v1", "arxiv_id": "1911.06009", "doi": "10.1109/TNNLS.2015.2400448", "citation_count": 28, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.3656}
{"id": "e59914809b7f139dde0bc314335f81c65f4e153f7b69a241b1645b97d01e107f", "sources": ["arxiv", "semantic_scholar"], "title": "Multivariate Uncertainty in Deep Learning", "abstract": "Deep learning has the potential to dramatically impact navigation and tracking state estimation problems critical to autonomous vehicles and robotics. Measurement uncertainties in state estimation systems based on Kalman and other Bayes filters are typically assumed to be a fixed covariance matrix. This assumption is risky, particularly for \"black box\" deep learning models, in which uncertainty can vary dramatically and unexpectedly. Accurate quantification of multivariate uncertainty will allow for the full potential of deep learning to be used more safely and reliably in these applications. We show how to model multivariate uncertainty for regression problems with neural networks, incorporating both aleatoric and epistemic sources of heteroscedastic uncertainty. We train a deep uncertainty covariance matrix model in two ways: directly using a multivariate Gaussian density loss function, and indirectly using end-to-end training through a Kalman filter. We experimentally show in a visual tracking problem the large impact that accurate multivariate uncertainty quantification can have on Kalman filter performance for both in-domain and out-of-domain evaluation data. We additionally show in a challenging visual odometry problem how end-to-end filter training can allow uncertainty predictions to compensate for filter weaknesses.", "authors": ["Rebecca L. Russell", "Christopher Reale"], "categories": ["cs.LG", "cs.NE", "cs.RO", "stat.ML"], "fields_of_study": ["Computer Science", "Medicine", "Mathematics"], "published_date": "2019-10-31", "url": "https://arxiv.org/abs/1910.14215", "pdf_url": "https://arxiv.org/pdf/1910.14215v2", "arxiv_id": "1910.14215", "doi": "10.1109/TNNLS.2021.3086757", "citation_count": 89, "influential_citation_count": 5, "has_code": false, "code_url": null, "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.4886}
{"id": "2d57732f229a3c1df4373bc82a6ed9210e490d2eb172992c26e7af16d8e8aad0", "sources": ["arxiv", "semantic_scholar"], "title": "Adversarial Feature Alignment: Avoid Catastrophic Forgetting in Incremental Task Lifelong Learning", "abstract": "Human beings are able to master a variety of knowledge and skills with ongoing learning. By contrast, dramatic performance degradation is observed when new tasks are added to an existing neural network model. This phenomenon, termed as \\emph{Catastrophic Forgetting}, is one of the major roadblocks that prevent deep neural networks from achieving human-level artificial intelligence. Several research efforts, e.g. \\emph{Lifelong} or \\emph{Continual} learning algorithms, have been proposed to tackle this problem. However, they either suffer from an accumulating drop in performance as the task sequence grows longer, or require to store an excessive amount of model parameters for historical memory, or cannot obtain competitive performance on the new tasks. In this paper, we focus on the incremental multi-task image classification scenario. Inspired by the learning process of human students, where they usually decompose complex tasks into easier goals, we propose an adversarial feature alignment method to avoid catastrophic forgetting. In our design, both the low-level visual features and high-level semantic features serve as soft targets and guide the training process in multiple stages, which provide sufficient supervised information of the old tasks and help to reduce forgetting. Due to the knowledge distillation and regularization phenomenons, the proposed method gains even better performance than finetuning on the new tasks, which makes it stand out from other methods. Extensive experiments in several typical lifelong learning scenarios demonstrate that our method outperforms the state-of-the-art methods in both accuracies on new tasks and performance preservation on old tasks.", "authors": ["Xin Yao", "Tianchi Huang", "Chenglei Wu", "Rui-Xiao Zhang", "Lifeng Sun"], "categories": ["cs.LG", "cs.CV", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics", "Medicine"], "published_date": "2019-10-24", "url": "https://arxiv.org/abs/1910.10986", "pdf_url": "https://arxiv.org/pdf/1910.10986v1", "arxiv_id": "1910.10986", "doi": "10.1162/neco_a_01232", "citation_count": 28, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "Neural Computation", "quality_score": 0.3656}
{"id": "14f88f161ee7d9229594fd3176cb8b4459975fd433c862cb17b66310adb3c91a", "sources": ["arxiv", "semantic_scholar"], "title": "Overcoming Forgetting in Federated Learning on Non-IID Data", "abstract": "We tackle the problem of Federated Learning in the non i.i.d. case, in which local models drift apart, inhibiting learning. Building on an analogy with Lifelong Learning, we adapt a solution for catastrophic forgetting to Federated Learning. We add a penalty term to the loss function, compelling all local models to converge to a shared optimum. We show that this can be done efficiently for communication (adding no further privacy risks), scaling with the number of nodes in the distributed setting. Our experiments show that this method is superior to competing ones for image recognition on the MNIST dataset.", "authors": ["Neta Shoham", "Tomer Avidor", "Aviv Keren", "Nadav Israel", "Daniel Benditkis", "Liron Mor-Yosef", "Itai Zeitak"], "categories": ["cs.LG", "cs.CR", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-10-17", "url": "https://arxiv.org/abs/1910.07796", "pdf_url": "https://arxiv.org/pdf/1910.07796v1", "arxiv_id": "1910.07796", "doi": null, "citation_count": 279, "influential_citation_count": 23, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.6901}
{"id": "13493d0c54fe6970dc46e14f4a4137485d644a1d69f546cdc931c2bbb8aff044", "sources": ["arxiv", "semantic_scholar"], "title": "Central Server Free Federated Learning over Single-sided Trust Social Networks", "abstract": "Federated learning has become increasingly important for modern machine learning, especially for data privacy-sensitive scenarios. Existing federated learning mostly adopts the central server-based architecture or centralized architecture. However, in many social network scenarios, centralized federated learning is not applicable (e.g., a central agent or server connecting all users may not exist, or the communication cost to the central server is not affordable). In this paper, we consider a generic setting: 1) the central server may not exist, and 2) the social network is unidirectional or of single-sided trust (i.e., user A trusts user B but user B may not trust user A). We propose a central server free federated learning algorithm, named Online Push-Sum (OPS) method, to handle this challenging but generic scenario. A rigorous regret analysis is also provided, which shows very interesting results on how users can benefit from communication with trusted users in the federated learning scenario. This work builds upon the fundamental algorithm framework and theoretical guarantees for federated learning in the generic social network scenario.", "authors": ["Chaoyang He", "Conghui Tan", "Hanlin Tang", "Shuang Qiu", "Ji Liu"], "categories": ["cs.LG", "cs.SI", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-10-11", "url": "https://arxiv.org/abs/1910.04956", "pdf_url": "https://arxiv.org/pdf/1910.04956v2", "arxiv_id": "1910.04956", "doi": null, "citation_count": 80, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4771}
{"id": "a61b41185182f77382877be7dfdd773d1cfa39516aa8a2f5919c40d94e7d0597", "sources": ["arxiv", "semantic_scholar"], "title": "Continual Learning Using Bayesian Neural Networks", "abstract": "Continual learning models allow to learn and adapt to new changes and tasks over time. However, in continual and sequential learning scenarios in which the models are trained using different data with various distributions, neural networks tend to forget the previously learned knowledge. This phenomenon is often referred to as catastrophic forgetting. The catastrophic forgetting is an inevitable problem in continual learning models for dynamic environments. To address this issue, we propose a method, called Continual Bayesian Learning Networks (CBLN), which enables the networks to allocate additional resources to adapt to new tasks without forgetting the previously learned tasks. Using a Bayesian Neural Network, CBLN maintains a mixture of Gaussian posterior distributions that are associated with different tasks. The proposed method tries to optimise the number of resources that are needed to learn each task and avoids an exponential increase in the number of resources that are involved in learning multiple tasks. The proposed method does not need to access the past training data and can choose suitable weights to classify the data points during the test time automatically based on an uncertainty criterion. We have evaluated our method on the MNIST and UCR time-series datasets. The evaluation results show that our method can address the catastrophic forgetting problem at a promising rate compared to the state-of-the-art models.", "authors": ["HongLin Li", "Payam Barnaghi", "Shirin Enshaeifar", "Frieder Ganz"], "categories": ["cs.LG", "cs.NE", "stat.ML"], "fields_of_study": ["Medicine", "Computer Science", "Mathematics"], "published_date": "2019-10-09", "url": "https://arxiv.org/abs/1910.04112", "pdf_url": "https://arxiv.org/pdf/1910.04112v2", "arxiv_id": "1910.04112", "doi": "10.1109/TNNLS.2020.3017292", "citation_count": 46, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.418}
{"id": "8e9b10efbc3967319a8a883b4af1b6f0d8ce49882904d393543d6c9c0f4d7ae7", "sources": ["arxiv", "semantic_scholar"], "title": "Continual Learning in Neural Networks", "abstract": "Artificial neural networks have exceeded human-level performance in accomplishing several individual tasks (e.g. voice recognition, object recognition, and video games). However, such success remains modest compared to human intelligence that can learn and perform an unlimited number of tasks. Humans' ability of learning and accumulating knowledge over their lifetime is an essential aspect of their intelligence. Continual machine learning aims at a higher level of machine intelligence through providing the artificial agents with the ability to learn online from a non-stationary and never-ending stream of data. A key component of such a never-ending learning process is to overcome the catastrophic forgetting of previously seen data, a problem that neural networks are well known to suffer from. The work described in this thesis has been dedicated to the investigation of continual learning and solutions to mitigate the forgetting phenomena in neural networks. To approach the continual learning problem, we first assume a task incremental setting where tasks are received one at a time and data from previous tasks are not stored. Since the task incremental setting can't be assumed in all continual learning scenarios, we also study the more general online continual setting. We consider an infinite stream of data drawn from a non-stationary distribution with a supervisory or self-supervisory training signal. The proposed methods in this thesis have tackled important aspects of continual learning. They were evaluated on different benchmarks and over various learning sequences. Advances in the state of the art of continual learning have been shown and challenges for bringing continual learning into application were critically identified.", "authors": ["Rahaf Aljundi"], "categories": ["cs.LG", "cs.CV", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-10-07", "url": "https://arxiv.org/abs/1910.02718", "pdf_url": "https://arxiv.org/pdf/1910.02718v2", "arxiv_id": "1910.02718", "doi": null, "citation_count": 48, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4225}
{"id": "fc92c4a0cdf116c55cbcccd5bd2309a7afa44770c3c9eca28f937f807f011955", "sources": ["arxiv", "semantic_scholar"], "title": "REMIND Your Neural Network to Prevent Catastrophic Forgetting", "abstract": "People learn throughout life. However, incrementally updating conventional neural networks leads to catastrophic forgetting. A common remedy is replay, which is inspired by how the brain consolidates memory. Replay involves fine-tuning a network on a mixture of new and old instances. While there is neuroscientific evidence that the brain replays compressed memories, existing methods for convolutional networks replay raw images. Here, we propose REMIND, a brain-inspired approach that enables efficient replay with compressed representations. REMIND is trained in an online manner, meaning it learns one example at a time, which is closer to how humans learn. Under the same constraints, REMIND outperforms other methods for incremental class learning on the ImageNet ILSVRC-2012 dataset. We probe REMIND's robustness to data ordering schemes known to induce catastrophic forgetting. We demonstrate REMIND's generality by pioneering online learning for Visual Question Answering (VQA).", "authors": ["Tyler L. Hayes", "Kushal Kafle", "Robik Shrestha", "Manoj Acharya", "Christopher Kanan"], "categories": ["cs.LG", "cs.CV", "cs.NE"], "fields_of_study": ["Computer Science"], "published_date": "2019-10-06", "url": "https://arxiv.org/abs/1910.02509", "pdf_url": "https://arxiv.org/pdf/1910.02509v3", "arxiv_id": "1910.02509", "doi": "10.1007/978-3-030-58598-3_28", "citation_count": 344, "influential_citation_count": 30, "has_code": false, "code_url": null, "venue": "European Conference on Computer Vision", "quality_score": 0.7457}
{"id": "8a2986c4747cac75e8edfade52725393f70bdc4266f411bc8539d09bc07143b9", "sources": ["arxiv", "semantic_scholar"], "title": "Multi-task Learning and Catastrophic Forgetting in Continual Reinforcement Learning", "abstract": "In this paper we investigate two hypothesis regarding the use of deep reinforcement learning in multiple tasks. The first hypothesis is driven by the question of whether a deep reinforcement learning algorithm, trained on two similar tasks, is able to outperform two single-task, individually trained algorithms, by more efficiently learning a new, similar task, that none of the three algorithms has encountered before. The second hypothesis is driven by the question of whether the same multi-task deep RL algorithm, trained on two similar tasks and augmented with elastic weight consolidation (EWC), is able to retain similar performance on the new task, as a similar algorithm without EWC, whilst being able to overcome catastrophic forgetting in the two previous tasks. We show that a multi-task Asynchronous Advantage Actor-Critic (GA3C) algorithm, trained on Space Invaders and Demon Attack, is in fact able to outperform two single-tasks GA3C versions, trained individually for each single-task, when evaluated on a new, third task, namely, Phoenix. We also show that, when training two trained multi-task GA3C algorithms on the third task, if one is augmented with EWC, it is not only able to achieve similar performance on the new task, but also capable of overcoming a substantial amount of catastrophic forgetting on the two previous tasks.", "authors": ["João Ribeiro", "Francisco S. Melo", "João Dias"], "categories": ["cs.LG", "cs.AI", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-09-22", "url": "https://arxiv.org/abs/1909.10008", "pdf_url": "https://arxiv.org/pdf/1909.10008v1", "arxiv_id": "1909.10008", "doi": "10.29007/g7bg", "citation_count": 12, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "Global Conference on Artificial Intelligence", "quality_score": 0.2785}
{"id": "44b6d5d08ad24ffb7f070fc11d433251e9cf549f90be589803eed1de396d6c08", "sources": ["arxiv", "semantic_scholar"], "title": "Adaptive Scheduling for Multi-Task Learning", "abstract": "To train neural machine translation models simultaneously on multiple tasks (languages), it is common to sample each task uniformly or in proportion to dataset sizes. As these methods offer little control over performance trade-offs, we explore different task scheduling approaches. We first consider existing non-adaptive techniques, then move on to adaptive schedules that over-sample tasks with poorer results compared to their respective baseline. As explicit schedules can be inefficient, especially if one task is highly over-sampled, we also consider implicit schedules, learning to scale learning rates or gradients of individual tasks instead. These techniques allow training multilingual models that perform better for low-resource language pairs (tasks with small amount of data), while minimizing negative effects on high-resource tasks.", "authors": ["Sébastien Jean", "Orhan Firat", "Melvin Johnson"], "categories": ["cs.LG", "cs.CL", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-09-13", "url": "https://arxiv.org/abs/1909.06434", "pdf_url": "https://arxiv.org/pdf/1909.06434v1", "arxiv_id": "1909.06434", "doi": null, "citation_count": 50, "influential_citation_count": 6, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4269}
{"id": "6f9d14bc8c84bdf818128463998dfb2596b28c44780bea95b6e27b9ec33bf498", "sources": ["arxiv", "semantic_scholar"], "title": "Efficient Continual Learning in Neural Networks with Embedding Regularization", "abstract": "Continual learning of deep neural networks is a key requirement for scaling them up to more complex applicative scenarios and for achieving real lifelong learning of these architectures. Previous approaches to the problem have considered either the progressive increase in the size of the networks, or have tried to regularize the network behavior to equalize it with respect to previously observed tasks. In the latter case, it is essential to understand what type of information best represents this past behavior. Common techniques include regularizing the past outputs, gradients, or individual weights. In this work, we propose a new, relatively simple and efficient method to perform continual learning by regularizing instead the network internal embeddings. To make the approach scalable, we also propose a dynamic sampling strategy to reduce the memory footprint of the required external storage. We show that our method performs favorably with respect to state-of-the-art approaches in the literature, while requiring significantly less space in memory and computational time. In addition, inspired inspired by to recent works, we evaluate the impact of selecting a more flexible model for the activation functions inside the network, evaluating the impact of catastrophic forgetting on the activation functions themselves.", "authors": ["Jary Pomponi", "Simone Scardapane", "Vincenzo Lomonaco", "Aurelio Uncini"], "categories": ["cs.LG", "cs.NE", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-09-09", "url": "https://arxiv.org/abs/1909.03742", "pdf_url": "https://arxiv.org/pdf/1909.03742v2", "arxiv_id": "1909.03742", "doi": "10.1016/j.neucom.2020.01.093", "citation_count": 47, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Neurocomputing", "quality_score": 0.4203}
{"id": "abc757b713c39fd22e7fa035f5cd86eac11d252e6f8957760f1876888b21f023", "sources": ["arxiv", "semantic_scholar"], "title": "Multi Pseudo Q-learning Based Deterministic Policy Gradient for Tracking Control of Autonomous Underwater Vehicles", "abstract": "This paper investigates trajectory tracking problem for a class of underactuated autonomous underwater vehicles (AUVs) with unknown dynamics and constrained inputs. Different from existing policy gradient methods which employ single actor-critic but cannot realize satisfactory tracking control accuracy and stable learning, our proposed algorithm can achieve high-level tracking control accuracy of AUVs and stable learning by applying a hybrid actors-critics architecture, where multiple actors and critics are trained to learn a deterministic policy and action-value function, respectively. Specifically, for the critics, the expected absolute Bellman error based updating rule is used to choose the worst critic to be updated in each time step. Subsequently, to calculate the loss function with more accurate target value for the chosen critic, Pseudo Q-learning, which uses sub-greedy policy to replace the greedy policy in Q-learning, is developed for continuous action spaces, and Multi Pseudo Q-learning (MPQ) is proposed to reduce the overestimation of action-value function and to stabilize the learning. As for the actors, deterministic policy gradient is applied to update the weights, and the final learned policy is defined as the average of all actors to avoid large but bad updates. Moreover, the stability analysis of the learning is given qualitatively. The effectiveness and generality of the proposed MPQ-based Deterministic Policy Gradient (MPQ-DPG) algorithm are verified by the application on AUV with two different reference trajectories. And the results demonstrate high-level tracking control accuracy and stable learning of MPQ-DPG. Besides, the results also validate that increasing the number of the actors and critics will further improve the performance.", "authors": ["Wenjie Shi", "Shiji Song", "Cheng Wu", "C. L. Philip Chen"], "categories": ["cs.LG", "cs.AI", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics", "Medicine"], "published_date": "2019-09-07", "url": "https://arxiv.org/abs/1909.03204", "pdf_url": "https://arxiv.org/pdf/1909.03204v1", "arxiv_id": "1909.03204", "doi": "10.1109/TNNLS.2018.2884797", "citation_count": 94, "influential_citation_count": 5, "has_code": false, "code_url": null, "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.4944}
{"id": "03e3ffa260e2a3066aea49e08f37996e8a312e3d3e8319551379818389eb4fae", "sources": ["arxiv", "semantic_scholar"], "title": "Lifelong Machine Learning with Deep Streaming Linear Discriminant Analysis", "abstract": "When an agent acquires new information, ideally it would immediately be capable of using that information to understand its environment. This is not possible using conventional deep neural networks, which suffer from catastrophic forgetting when they are incrementally updated, with new knowledge overwriting established representations. A variety of approaches have been developed that attempt to mitigate catastrophic forgetting in the incremental batch learning scenario, where a model learns from a series of large collections of labeled samples. However, in this setting, inference is only possible after a batch has been accumulated, which prohibits many applications. An alternative paradigm is online learning in a single pass through the training dataset on a resource constrained budget, which is known as streaming learning. Streaming learning has been much less studied in the deep learning community. In streaming learning, an agent learns instances one-by-one and can be tested at any time, rather than only after learning a large batch. Here, we revisit streaming linear discriminant analysis, which has been widely used in the data mining research community. By combining streaming linear discriminant analysis with deep learning, we are able to outperform both incremental batch learning and streaming learning algorithms on both ImageNet ILSVRC-2012 and CORe50, a dataset that involves learning to classify from temporally ordered samples.", "authors": ["Tyler L. Hayes", "Christopher Kanan"], "categories": ["cs.LG", "cs.CV", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-09-04", "url": "https://arxiv.org/abs/1909.01520", "pdf_url": "https://arxiv.org/pdf/1909.01520v3", "arxiv_id": "1909.01520", "doi": "10.1109/CVPRW50498.2020.00118", "citation_count": 171, "influential_citation_count": 26, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.7157}
{"id": "477cae8119d04c253a9b12501eeee95de703a623443aa4f205d511c38ef1d9c7", "sources": ["arxiv", "semantic_scholar"], "title": "A Neural Network for Semi-Supervised Learning on Manifolds", "abstract": "Semi-supervised learning algorithms typically construct a weighted graph of data points to represent a manifold. However, an explicit graph representation is problematic for neural networks operating in the online setting. Here, we propose a feed-forward neural network capable of semi-supervised learning on manifolds without using an explicit graph representation. Our algorithm uses channels that represent localities on the manifold such that correlations between channels represent manifold structure. The proposed neural network has two layers. The first layer learns to build a representation of low-dimensional manifolds in the input data as proposed recently in [8]. The second learns to classify data using both occasional supervision and similarity of the manifold representation of the data. The channel carrying label information for the second layer is assumed to be \"silent\" most of the time. Learning in both layers is Hebbian, making our network design biologically plausible. We experimentally demonstrate the effect of semi-supervised learning on non-trivial manifolds.", "authors": ["Alexander Genkin", "Anirvan M. Sengupta", "Dmitri Chklovskii"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-08-21", "url": "https://arxiv.org/abs/1908.08145", "pdf_url": "https://arxiv.org/pdf/1908.08145v1", "arxiv_id": "1908.08145", "doi": "10.1007/978-3-030-30487-4_30", "citation_count": 7, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "International Conference on Artificial Neural Networks", "quality_score": 0.2258}
{"id": "ff30fc95645ddf8c03b19e1e82ae8ff35eb866940fc912efa646547ad2517f78", "sources": ["arxiv", "semantic_scholar"], "title": "Weight Friction: A Simple Method to Overcome Catastrophic Forgetting and Enable Continual Learning", "abstract": "In recent years, deep neural networks have found success in replicating human-level cognitive skills, yet they suffer from several major obstacles. One significant limitation is the inability to learn new tasks without forgetting previously learned tasks, a shortcoming known as catastrophic forgetting. In this research, we propose a simple method to overcome catastrophic forgetting and enable continual learning in neural networks. We draw inspiration from principles in neurology and physics to develop the concept of weight friction. Weight friction operates by a modification to the update rule in the gradient descent optimization method. It converges at a rate comparable to that of the stochastic gradient descent algorithm and can operate over multiple task domains. It performs comparably to current methods while offering improvements in computation and memory efficiency.", "authors": ["Gabrielle K. Liu"], "categories": ["cs.LG", "cs.NE", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-08-02", "url": "https://arxiv.org/abs/1908.01052", "pdf_url": "https://arxiv.org/pdf/1908.01052v2", "arxiv_id": "1908.01052", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.0}
{"id": "3567598232071ba9a5c9b13c2688d436401d99bb0c0b4c879add1693a2ed3561", "sources": ["arxiv", "semantic_scholar"], "title": "Toward Understanding Catastrophic Forgetting in Continual Learning", "abstract": "We study the relationship between catastrophic forgetting and properties of task sequences. In particular, given a sequence of tasks, we would like to understand which properties of this sequence influence the error rates of continual learning algorithms trained on the sequence. To this end, we propose a new procedure that makes use of recent developments in task space modeling as well as correlation analysis to specify and analyze the properties we are interested in. As an application, we apply our procedure to study two properties of a task sequence: (1) total complexity and (2) sequential heterogeneity. We show that error rates are strongly and positively correlated to a task sequence's total complexity for some state-of-the-art algorithms. We also show that, surprisingly, the error rates have no or even negative correlations in some cases to sequential heterogeneity. Our findings suggest directions for improving continual learning benchmarks and methods.", "authors": ["Cuong V. Nguyen", "Alessandro Achille", "Michael Lam", "Tal Hassner", "Vijay Mahadevan", "Stefano Soatto"], "categories": ["cs.LG", "cs.CV", "stat.ML"], "fields_of_study": ["Mathematics", "Computer Science"], "published_date": "2019-08-02", "url": "https://arxiv.org/abs/1908.01091", "pdf_url": "https://arxiv.org/pdf/1908.01091v1", "arxiv_id": "1908.01091", "doi": null, "citation_count": 124, "influential_citation_count": 3, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.5242}
{"id": "a5e8f8fbbd2cf8d3e3d73016ce73b2f728715ab1f3c6d060273491bd1b08796a", "sources": ["arxiv", "semantic_scholar"], "title": "An Optimistic Perspective on Offline Reinforcement Learning", "abstract": "Off-policy reinforcement learning (RL) using a fixed offline dataset of logged interactions is an important consideration in real world applications. This paper studies offline RL using the DQN replay dataset comprising the entire replay experience of a DQN agent on 60 Atari 2600 games. We demonstrate that recent off-policy deep RL algorithms, even when trained solely on this fixed dataset, outperform the fully trained DQN agent. To enhance generalization in the offline setting, we present Random Ensemble Mixture (REM), a robust Q-learning algorithm that enforces optimal Bellman consistency on random convex combinations of multiple Q-value estimates. Offline REM trained on the DQN replay dataset surpasses strong RL baselines. Ablation studies highlight the role of offline dataset size and diversity as well as the algorithm choice in our positive results. Overall, the results here present an optimistic view that robust RL algorithms trained on sufficiently large and diverse offline datasets can lead to high quality policies. The DQN replay dataset can serve as an offline RL benchmark and is open-sourced.", "authors": ["Rishabh Agarwal", "Dale Schuurmans", "Mohammad Norouzi"], "categories": ["cs.LG", "cs.AI", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-07-10", "url": "https://arxiv.org/abs/1907.04543", "pdf_url": "https://arxiv.org/pdf/1907.04543v4", "arxiv_id": "1907.04543", "doi": null, "citation_count": 71, "influential_citation_count": 12, "has_code": true, "code_url": null, "venue": "arXiv.org", "quality_score": 0.557}
{"id": "d78b8ac18b1695bf53deefea1e6d51267615fe40c1c5679b44d7037003b81c56", "sources": ["arxiv", "semantic_scholar"], "title": "On-Policy Robot Imitation Learning from a Converging Supervisor", "abstract": "Existing on-policy imitation learning algorithms, such as DAgger, assume access to a fixed supervisor. However, there are many settings where the supervisor may evolve during policy learning, such as a human performing a novel task or an improving algorithmic controller. We formalize imitation learning from a \"converging supervisor\" and provide sublinear static and dynamic regret guarantees against the best policy in hindsight with labels from the converged supervisor, even when labels during learning are only from intermediate supervisors. We then show that this framework is closely connected to a class of reinforcement learning (RL) algorithms known as dual policy iteration (DPI), which alternate between training a reactive learner with imitation learning and a model-based supervisor with data from the learner. Experiments suggest that when this framework is applied with the state-of-the-art deep model-based RL algorithm PETS as an improving supervisor, it outperforms deep RL baselines on continuous control tasks and provides up to an 80-fold speedup in policy evaluation.", "authors": ["Ashwin Balakrishna", "Brijen Thananjeyan", "Jonathan Lee", "Felix Li", "Arsh Zahed", "Joseph E. Gonzalez", "Ken Goldberg"], "categories": ["cs.LG", "cs.AI", "cs.RO"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-07-08", "url": "https://arxiv.org/abs/1907.03423", "pdf_url": "https://arxiv.org/pdf/1907.03423v7", "arxiv_id": "1907.03423", "doi": null, "citation_count": 20, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "Conference on Robot Learning", "quality_score": 0.3306}
{"id": "f8b81b6f2c776ada28e6c33c59eae28e98daac557679b9eacf0d63357d17bfb4", "sources": ["arxiv", "semantic_scholar"], "title": "Further advantages of data augmentation on convolutional neural networks", "abstract": "Data augmentation is a popular technique largely used to enhance the training of convolutional neural networks. Although many of its benefits are well known by deep learning researchers and practitioners, its implicit regularization effects, as compared to popular explicit regularization techniques, such as weight decay and dropout, remain largely unstudied. As a matter of fact, convolutional neural networks for image object classification are typically trained with both data augmentation and explicit regularization, assuming the benefits of all techniques are complementary. In this paper, we systematically analyze these techniques through ablation studies of different network architectures trained with different amounts of training data. Our results unveil a largely ignored advantage of data augmentation: networks trained with just data augmentation more easily adapt to different architectures and amount of training data, as opposed to weight decay and dropout, which require specific fine-tuning of their hyperparameters.", "authors": ["Alex Hernández-García", "Peter König"], "categories": ["cs.CV", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2019-06-26", "url": "https://arxiv.org/abs/1906.11052", "pdf_url": "https://arxiv.org/pdf/1906.11052v1", "arxiv_id": "1906.11052", "doi": "10.1007/978-3-030-01418-6_10", "citation_count": 125, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "International Conference on Artificial Neural Networks", "quality_score": 0.5251}
{"id": "d80dc7a957e980b6b5dbcf5829f2e1750c8b8de4929a3ac6d79cd8b9c22a57fc", "sources": ["arxiv", "semantic_scholar"], "title": "Evolutionary Reinforcement Learning for Sample-Efficient Multiagent Coordination", "abstract": "Many cooperative multiagent reinforcement learning environments provide agents with a sparse team-based reward, as well as a dense agent-specific reward that incentivizes learning basic skills. Training policies solely on the team-based reward is often difficult due to its sparsity. Furthermore, relying solely on the agent-specific reward is sub-optimal because it usually does not capture the team coordination objective. A common approach is to use reward shaping to construct a proxy reward by combining the individual rewards. However, this requires manual tuning for each environment. We introduce Multiagent Evolutionary Reinforcement Learning (MERL), a split-level training platform that handles the two objectives separately through two optimization processes. An evolutionary algorithm maximizes the sparse team-based objective through neuroevolution on a population of teams. Concurrently, a gradient-based optimizer trains policies to only maximize the dense agent-specific rewards. The gradient-based policies are periodically added to the evolutionary population as a way of information transfer between the two optimization processes. This enables the evolutionary algorithm to use skills learned via the agent-specific rewards toward optimizing the global objective. Results demonstrate that MERL significantly outperforms state-of-the-art methods, such as MADDPG, on a number of difficult coordination benchmarks.", "authors": ["Shauharda Khadka", "Somdeb Majumdar", "Santiago Miret", "Stephen McAleer", "Kagan Tumer"], "categories": ["cs.LG", "cs.AI", "cs.MA", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-06-18", "url": "https://arxiv.org/abs/1906.07315", "pdf_url": "https://arxiv.org/pdf/1906.07315v3", "arxiv_id": "1906.07315", "doi": null, "citation_count": 70, "influential_citation_count": 5, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.4628}
{"id": "8eaca0e7208827a3bd1cf0e5311b157735dea62c0816bc927e17fa9da7230792", "sources": ["arxiv", "semantic_scholar"], "title": "Iterative Model-Based Reinforcement Learning Using Simulations in the Differentiable Neural Computer", "abstract": "We propose a lifelong learning architecture, the Neural Computer Agent (NCA), where a Reinforcement Learning agent is paired with a predictive model of the environment learned by a Differentiable Neural Computer (DNC). The agent and DNC model are trained in conjunction iteratively. The agent improves its policy in simulations generated by the DNC model and rolls out the policy to the live environment, collecting experiences in new portions or tasks of the environment for further learning. Experiments in two synthetic environments show that DNC models can continually learn from pixels alone to simulate new tasks as they are encountered by the agent, while the agents can be successfully trained to solve the tasks using Proximal Policy Optimization entirely in simulations.", "authors": ["Adeel Mufti", "Svetlin Penkov", "Subramanian Ramamoorthy"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Mathematics", "Computer Science"], "published_date": "2019-06-17", "url": "https://arxiv.org/abs/1906.07248", "pdf_url": "https://arxiv.org/pdf/1906.07248v1", "arxiv_id": "1906.07248", "doi": null, "citation_count": 2, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1193}
{"id": "0140d04ce0a94eb41b0911c3a9caf1ea6bd8202225cf3ea27b5ffae23bd61323", "sources": ["arxiv", "semantic_scholar"], "title": "Conditional Computation for Continual Learning", "abstract": "Catastrophic forgetting of connectionist neural networks is caused by the global sharing of parameters among all training examples. In this study, we analyze parameter sharing under the conditional computation framework where the parameters of a neural network are conditioned on each input example. At one extreme, if each input example uses a disjoint set of parameters, there is no sharing of parameters thus no catastrophic forgetting. At the other extreme, if the parameters are the same for every example, it reduces to the conventional neural network. We then introduce a clipped version of maxout networks which lies in the middle, i.e. parameters are shared partially among examples. Based on the parameter sharing analysis, we can locate a limited set of examples that are interfered when learning a new example. We propose to perform rehearsal on this set to prevent forgetting, which is termed as conditional rehearsal. Finally, we demonstrate the effectiveness of the proposed method in an online non-stationary setup, where updates are made after each new example and the distribution of the received example shifts over time.", "authors": ["Min Lin", "Jie Fu", "Yoshua Bengio"], "categories": ["cs.LG", "cs.NE", "stat.ML"], "fields_of_study": ["Mathematics", "Computer Science"], "published_date": "2019-06-16", "url": "https://arxiv.org/abs/1906.06635", "pdf_url": "https://arxiv.org/pdf/1906.06635v1", "arxiv_id": "1906.06635", "doi": null, "citation_count": 11, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2698}
{"id": "c1afb838d81b614ad509b95ed19398563121fc5eda6f9592f1a83f3ce89eaa02", "sources": ["arxiv", "semantic_scholar"], "title": "Multiple instance learning with graph neural networks", "abstract": "Multiple instance learning (MIL) aims to learn the mapping between a bag of instances and the bag-level label. In this paper, we propose a new end-to-end graph neural network (GNN) based algorithm for MIL: we treat each bag as a graph and use GNN to learn the bag embedding, in order to explore the useful structural information among instances in bags. The final graph representation is fed into a classifier for label prediction. Our algorithm is the first attempt to use GNN for MIL. We empirically show that the proposed algorithm achieves the state of the art performance on several popular MIL data sets without losing model interpretability.", "authors": ["Ming Tu", "Jing Huang", "Xiaodong He", "Bowen Zhou"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-06-12", "url": "https://arxiv.org/abs/1906.04881", "pdf_url": "https://arxiv.org/pdf/1906.04881v1", "arxiv_id": "1906.04881", "doi": null, "citation_count": 70, "influential_citation_count": 12, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.557}
{"id": "7db323e2fc033d96a248fdf2ad83f72327b875426546d37fa625cdfec4ff0ad0", "sources": ["arxiv", "semantic_scholar"], "title": "Continual Reinforcement Learning deployed in Real-life using Policy Distillation and Sim2Real Transfer", "abstract": "We focus on the problem of teaching a robot to solve tasks presented sequentially, i.e., in a continual learning scenario. The robot should be able to solve all tasks it has encountered, without forgetting past tasks. We provide preliminary work on applying Reinforcement Learning to such setting, on 2D navigation tasks for a 3 wheel omni-directional robot. Our approach takes advantage of state representation learning and policy distillation. Policies are trained using learned features as input, rather than raw observations, allowing better sample efficiency. Policy distillation is used to combine multiple policies into a single one that solves all encountered tasks.", "authors": ["René Traoré", "Hugo Caselles-Dupré", "Timothée Lesort", "Te Sun", "Natalia Díaz-Rodríguez", "David Filliat"], "categories": ["cs.LG", "cs.RO", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-06-11", "url": "https://arxiv.org/abs/1906.04452", "pdf_url": "https://arxiv.org/pdf/1906.04452v1", "arxiv_id": "1906.04452", "doi": null, "citation_count": 47, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4203}
{"id": "0e7d68aeb018f41fa37ad9c8a49bdb8fc7a14bd86cf3129f350790ad9959cfcd", "sources": ["arxiv", "semantic_scholar"], "title": "Meta-Learning Neural Bloom Filters", "abstract": "There has been a recent trend in training neural networks to replace data structures that have been crafted by hand, with an aim for faster execution, better accuracy, or greater compression. In this setting, a neural data structure is instantiated by training a network over many epochs of its inputs until convergence. In applications where inputs arrive at high throughput, or are ephemeral, training a network from scratch is not practical. This motivates the need for few-shot neural data structures. In this paper we explore the learning of approximate set membership over a set of data in one-shot via meta-learning. We propose a novel memory architecture, the Neural Bloom Filter, which is able to achieve significant compression gains over classical Bloom Filters and existing memory-augmented neural networks.", "authors": ["Jack W Rae", "Sergey Bartunov", "Timothy P Lillicrap"], "categories": ["cs.LG", "cs.DB", "cs.DS", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-06-10", "url": "https://arxiv.org/abs/1906.04304", "pdf_url": "https://arxiv.org/pdf/1906.04304v1", "arxiv_id": "1906.04304", "doi": null, "citation_count": 40, "influential_citation_count": 4, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.4032}
{"id": "3db256f66d7fbbb49a31d00b23ad59c974c1c4befab02172eef534682133a18a", "sources": ["arxiv", "semantic_scholar"], "title": "Localizing Catastrophic Forgetting in Neural Networks", "abstract": "Artificial neural networks (ANNs) suffer from catastrophic forgetting when trained on a sequence of tasks. While this phenomenon was studied in the past, there is only very limited recent research on this phenomenon. We propose a method for determining the contribution of individual parameters in an ANN to catastrophic forgetting. The method is used to analyze an ANNs response to three different continual learning scenarios.", "authors": ["Felix Wiewel", "Bin Yang"], "categories": ["cs.LG", "cs.AI", "cs.NE", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-06-06", "url": "https://arxiv.org/abs/1906.02568", "pdf_url": "https://arxiv.org/pdf/1906.02568v1", "arxiv_id": "1906.02568", "doi": null, "citation_count": 13, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2865}
{"id": "3c562f25e6118af360dad35fa249ef155ce0b8413a1cc37e5add0b42b204ab71", "sources": ["arxiv", "semantic_scholar"], "title": "Uncertainty-guided Continual Learning with Bayesian Neural Networks", "abstract": "Continual learning aims to learn new tasks without forgetting previously learned ones. This is especially challenging when one cannot access data from previous tasks and when the model has a fixed capacity. Current regularization-based continual learning algorithms need an external representation and extra computation to measure the parameters' \\textit{importance}. In contrast, we propose Uncertainty-guided Continual Bayesian Neural Networks (UCB), where the learning rate adapts according to the uncertainty defined in the probability distribution of the weights in networks. Uncertainty is a natural way to identify \\textit{what to remember} and \\textit{what to change} as we continually learn, and thus mitigate catastrophic forgetting. We also show a variant of our model, which uses uncertainty for weight pruning and retains task performance after pruning by saving binary masks per tasks. We evaluate our UCB approach extensively on diverse object classification datasets with short and long sequences of tasks and report superior or on-par performance compared to existing approaches. Additionally, we show that our model does not necessarily need task information at test time, i.e. it does not presume knowledge of which task a sample belongs to.", "authors": ["Sayna Ebrahimi", "Mohamed Elhoseiny", "Trevor Darrell", "Marcus Rohrbach"], "categories": ["cs.LG", "cs.AI", "cs.CV", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-06-06", "url": "https://arxiv.org/abs/1906.02425", "pdf_url": "https://arxiv.org/pdf/1906.02425v2", "arxiv_id": "1906.02425", "doi": null, "citation_count": 213, "influential_citation_count": 17, "has_code": false, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.6276}
{"id": "805cf18e1f05a9ed7d97cf87630192ac8f376ac184ccb0d8e5c0e045068b9642", "sources": ["arxiv", "semantic_scholar"], "title": "Gradient-Based Neural DAG Learning", "abstract": "We propose a novel score-based approach to learning a directed acyclic graph (DAG) from observational data. We adapt a recently proposed continuous constrained optimization formulation to allow for nonlinear relationships between variables using neural networks. This extension allows to model complex interactions while avoiding the combinatorial nature of the problem. In addition to comparing our method to existing continuous optimization methods, we provide missing empirical comparisons to nonlinear greedy search methods. On both synthetic and real-world data sets, this new method outperforms current continuous methods on most tasks, while being competitive with existing greedy search methods on important metrics for causal inference.", "authors": ["Sébastien Lachapelle", "Philippe Brouillard", "Tristan Deleu", "Simon Lacoste-Julien"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-06-05", "url": "https://arxiv.org/abs/1906.02226", "pdf_url": "https://arxiv.org/pdf/1906.02226v2", "arxiv_id": "1906.02226", "doi": null, "citation_count": 349, "influential_citation_count": 54, "has_code": false, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.8702}
{"id": "1b44c1ff74dedd2086398d0d386d20c3121016322c219ada046e1d97f24e548c", "sources": ["arxiv", "semantic_scholar"], "title": "Regression Networks for Meta-Learning Few-Shot Classification", "abstract": "We propose regression networks for the problem of few-shot classification, where a classifier must generalize to new classes not seen in the training set, given only a small number of examples of each class. In high dimensional embedding spaces the direction of data generally contains richer information than magnitude. Next to this, state-of-the-art few-shot metric methods that compare distances with aggregated class representations, have shown superior performance. Combining these two insights, we propose to meta-learn classification of embedded points by regressing the closest approximation in every class subspace while using the regression error as a distance metric. Similarly to recent approaches for few-shot learning, regression networks reflect a simple inductive bias that is beneficial in this limited-data regime and they achieve excellent results, especially when more aggregate class representations can be formed with multiple shots.", "authors": ["Arnout Devos", "Matthias Grossglauser"], "categories": ["cs.LG", "cs.CV", "stat.ML"], "fields_of_study": ["Mathematics", "Computer Science"], "published_date": "2019-05-31", "url": "https://arxiv.org/abs/1905.13613", "pdf_url": "https://arxiv.org/pdf/1905.13613v2", "arxiv_id": "1905.13613", "doi": null, "citation_count": 12, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2785}
{"id": "c0a3dd1441ee82a10b7eb9e2ed72cb0ab1ff9164bb9c85b64d77f33d3717bb88", "sources": ["arxiv", "semantic_scholar"], "title": "Meta-Learning Representations for Continual Learning", "abstract": "A continual learning agent should be able to build on top of existing knowledge to learn on new data quickly while minimizing forgetting. Current intelligent systems based on neural network function approximators arguably do the opposite---they are highly prone to forgetting and rarely trained to facilitate future learning. One reason for this poor behavior is that they learn from a representation that is not explicitly trained for these two goals. In this paper, we propose OML, an objective that directly minimizes catastrophic interference by learning representations that accelerate future learning and are robust to forgetting under online updates in continual learning. We show that it is possible to learn naturally sparse representations that are more effective for online updating. Moreover, our algorithm is complementary to existing continual learning strategies, such as MER and GEM. Finally, we demonstrate that a basic online updating strategy on representations learned by OML is competitive with rehearsal based methods for continual learning. We release an implementation of our method at https://github.com/khurramjaved96/mrcl .", "authors": ["Khurram Javed", "Martha White"], "categories": ["cs.LG", "cs.AI", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-05-29", "url": "https://arxiv.org/abs/1905.12588", "pdf_url": "https://arxiv.org/pdf/1905.12588v2", "arxiv_id": "1905.12588", "doi": null, "citation_count": 378, "influential_citation_count": 48, "has_code": true, "code_url": "https://github.com/khurramjaved96/mrcl", "venue": "Neural Information Processing Systems", "quality_score": 0.8451}
{"id": "4f9d7b448124f870c3893d508264983a20fc17975bbaa34a8cbba53a740f9779", "sources": ["arxiv", "semantic_scholar"], "title": "An Inertial Newton Algorithm for Deep Learning", "abstract": "We introduce a new second-order inertial optimization method for machine learning called INNA. It exploits the geometry of the loss function while only requiring stochastic approximations of the function values and the generalized gradients. This makes INNA fully implementable and adapted to large-scale optimization problems such as the training of deep neural networks. The algorithm combines both gradient-descent and Newton-like behaviors as well as inertia. We prove the convergence of INNA for most deep learning problems. To do so, we provide a well-suited framework to analyze deep learning loss functions involving tame optimization in which we study a continuous dynamical system together with its discrete stochastic approximations. We prove sublinear convergence for the continuous-time differential inclusion which underlies our algorithm. Additionally, we also show how standard optimization mini-batch methods applied to non-smooth non-convex problems can yield a certain type of spurious stationary points never discussed before. We address this issue by providing a theoretical framework around the new idea of $D$-criticality; we then give a simple asymptotic analysis of INNA. Our algorithm allows for using an aggressive learning rate of $o(1/\\log k)$. From an empirical viewpoint, we show that INNA returns competitive results with respect to state of the art (stochastic gradient descent, ADAGRAD, ADAM) on popular deep learning benchmark problems.", "authors": ["Camille Castera", "Jérôme Bolte", "Cédric Févotte", "Edouard Pauwels"], "categories": ["cs.LG", "math.OC", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-05-29", "url": "https://arxiv.org/abs/1905.12278", "pdf_url": "https://arxiv.org/pdf/1905.12278v6", "arxiv_id": "1905.12278", "doi": null, "citation_count": 74, "influential_citation_count": 12, "has_code": false, "code_url": null, "venue": "Journal of machine learning research", "quality_score": 0.557}
{"id": "acbdc6955f5f8fc6032d56836cd4dbe97e64c06fe133f54815c0dd5ad7dbd9ce", "sources": ["arxiv", "semantic_scholar"], "title": "Graph Learning Network: A Structure Learning Algorithm", "abstract": "Recently, graph neural networks (GNNs) have proved to be suitable in tasks on unstructured data. Particularly in tasks as community detection, node classification, and link prediction. However, most GNN models still operate with static relationships. We propose the Graph Learning Network (GLN), a simple yet effective process to learn node embeddings and structure prediction functions. Our model uses graph convolutions to propose expected node features, and predict the best structure based on them. We repeat these steps recursively to enhance the prediction and the embeddings.", "authors": ["Darwin Saire Pilco", "Adín Ramírez Rivera"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-05-29", "url": "https://arxiv.org/abs/1905.12665", "pdf_url": "https://arxiv.org/pdf/1905.12665v3", "arxiv_id": "1905.12665", "doi": null, "citation_count": 17, "influential_citation_count": 0, "has_code": true, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3138}
{"id": "5e46f3c1191b5d73d174b1a3f7269ec97ae1741f395ea0428e7b6544a9b9d7a0", "sources": ["arxiv", "semantic_scholar"], "title": "Unified Probabilistic Deep Continual Learning through Generative Replay and Open Set Recognition", "abstract": "Modern deep neural networks are well known to be brittle in the face of unknown data instances and recognition of the latter remains a challenge. Although it is inevitable for continual-learning systems to encounter such unseen concepts, the corresponding literature appears to nonetheless focus primarily on alleviating catastrophic interference with learned representations. In this work, we introduce a probabilistic approach that connects these perspectives based on variational inference in a single deep autoencoder model. Specifically, we propose to bound the approximate posterior by fitting regions of high density on the basis of correctly classified data points. These bounds are shown to serve a dual purpose: unseen unknown out-of-distribution data can be distinguished from already trained known tasks towards robust application. Simultaneously, to retain already acquired knowledge, a generative replay process can be narrowed to strictly in-distribution samples, in order to significantly alleviate catastrophic interference.", "authors": ["Martin Mundt", "Iuliia Pliushch", "Sagnik Majumder", "Yongwon Hong", "Visvanathan Ramesh"], "categories": ["cs.LG", "cs.CV", "cs.NE", "stat.ML"], "fields_of_study": ["Computer Science", "Medicine", "Mathematics"], "published_date": "2019-05-28", "url": "https://arxiv.org/abs/1905.12019", "pdf_url": "https://arxiv.org/pdf/1905.12019v5", "arxiv_id": "1905.12019", "doi": "10.3390/jimaging8040093", "citation_count": 43, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "Journal of Imaging", "quality_score": 0.4109}
{"id": "e35dc4901a63db2a08afef43d6010b25f7b7f8543cc4c7ff4d6ebfbf07d994de", "sources": ["arxiv", "semantic_scholar"], "title": "Lifelong Neural Predictive Coding: Learning Cumulatively Online without Forgetting", "abstract": "In lifelong learning systems based on artificial neural networks, one of the biggest obstacles is the inability to retain old knowledge as new information is encountered. This phenomenon is known as catastrophic forgetting. In this paper, we propose a new kind of connectionist architecture, the Sequential Neural Coding Network, that is robust to forgetting when learning from streams of data points and, unlike networks of today, does not learn via the popular back-propagation of errors. Grounded in the neurocognitive theory of predictive processing, our model adapts synapses in a biologically-plausible fashion while another neural system learns to direct and control this cortex-like structure, mimicking some of the task-executive control functionality of the basal ganglia. In our experiments, we demonstrate that our self-organizing system experiences significantly less forgetting compared to standard neural models, outperforming a swath of previously proposed methods, including rehearsal/data buffer-based methods, on both standard (SplitMNIST, Split Fashion MNIST, etc.) and custom benchmarks even though it is trained in a stream-like fashion. Our work offers evidence that emulating mechanisms in real neuronal systems, e.g., local learning, lateral competition, can yield new directions and possibilities for tackling the grand challenge of lifelong machine learning.", "authors": ["Alexander Ororbia", "Ankur Mali", "Daniel Kifer", "C. Lee Giles"], "categories": ["cs.LG", "cs.NE", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-05-25", "url": "https://arxiv.org/abs/1905.10696", "pdf_url": "https://arxiv.org/pdf/1905.10696v4", "arxiv_id": "1905.10696", "doi": "10.52202/068431-0425", "citation_count": 22, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "Neural Information Processing Systems", "quality_score": 0.3404}
{"id": "f181c7032c36bb0ea1caa42f0606f887075fed3eff7b64167bc622bd20aad865", "sources": ["arxiv", "semantic_scholar"], "title": "Automatic Machine Learning by Pipeline Synthesis using Model-Based Reinforcement Learning and a Grammar", "abstract": "Automatic machine learning is an important problem in the forefront of machine learning. The strongest AutoML systems are based on neural networks, evolutionary algorithms, and Bayesian optimization. Recently AlphaD3M reached state-of-the-art results with an order of magnitude speedup using reinforcement learning with self-play. In this work we extend AlphaD3M by using a pipeline grammar and a pre-trained model which generalizes from many different datasets and similar tasks. Our results demonstrate improved performance compared with our earlier work and existing methods on AutoML benchmark datasets for classification and regression tasks. In the spirit of reproducible research we make our data, models, and code publicly available.", "authors": ["Iddo Drori", "Yamuna Krishnamurthy", "Raoni Lourenco", "Remi Rampin", "Kyunghyun Cho", "Claudio Silva", "Juliana Freire"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-05-24", "url": "https://arxiv.org/abs/1905.10345", "pdf_url": "https://arxiv.org/pdf/1905.10345v1", "arxiv_id": "1905.10345", "doi": null, "citation_count": 30, "influential_citation_count": 6, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4225}
{"id": "0eeb73a3264989b1be48eae00b199f180fd3962dd05f621f0cbdb5c803f7c107", "sources": ["arxiv", "semantic_scholar"], "title": "A comprehensive, application-oriented study of catastrophic forgetting in DNNs", "abstract": "We present a large-scale empirical study of catastrophic forgetting (CF) in modern Deep Neural Network (DNN) models that perform sequential (or: incremental) learning. A new experimental protocol is proposed that enforces typical constraints encountered in application scenarios. As the investigation is empirical, we evaluate CF behavior on the hitherto largest number of visual classification datasets, from each of which we construct a representative number of Sequential Learning Tasks (SLTs) in close alignment to previous works on CF. Our results clearly indicate that there is no model that avoids CF for all investigated datasets and SLTs under application conditions. We conclude with a discussion of potential solutions and workarounds to CF, notably for the EWC and IMM models.", "authors": ["B. Pfülb", "A. Gepperth"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-05-20", "url": "https://arxiv.org/abs/1905.08101", "pdf_url": "https://arxiv.org/pdf/1905.08101v1", "arxiv_id": "1905.08101", "doi": null, "citation_count": 95, "influential_citation_count": 4, "has_code": false, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.4956}
{"id": "eadf6c9c394be4bddd136cbdd22aeb4bfa060d637186ae3491a97dc68738b88b", "sources": ["arxiv", "semantic_scholar"], "title": "Catastrophic forgetting: still a problem for DNNs", "abstract": "We investigate the performance of DNNs when trained on class-incremental visual problems consisting of initial training, followed by retraining with added visual classes. Catastrophic forgetting (CF) behavior is measured using a new evaluation procedure that aims at an application-oriented view of incremental learning. In particular, it imposes that model selection must be performed on the initial dataset alone, as well as demanding that retraining control be performed only using the retraining dataset, as initial dataset is usually too large to be kept. Experiments are conducted on class-incremental problems derived from MNIST, using a variety of different DNN models, some of them recently proposed to avoid catastrophic forgetting. When comparing our new evaluation procedure to previous approaches for assessing CF, we find their findings are completely negated, and that none of the tested methods can avoid CF in all experiments. This stresses the importance of a realistic empirical measurement procedure for catastrophic forgetting, and the need for further research in incremental learning for DNNs.", "authors": ["B. Pfülb", "A. Gepperth", "S. Abdullah", "A. Kilian"], "categories": ["cs.LG", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2019-05-20", "url": "https://arxiv.org/abs/1905.08077", "pdf_url": "https://arxiv.org/pdf/1905.08077v1", "arxiv_id": "1905.08077", "doi": "10.1007/978-3-030-01418-6_48", "citation_count": 25, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "International Conference on Artificial Neural Networks", "quality_score": 0.3537}
{"id": "e8c445b9d0c94a1430d378f7aac1993bc592d155ef21abe15d11589df56794aa", "sources": ["arxiv", "semantic_scholar"], "title": "Alpha MAML: Adaptive Model-Agnostic Meta-Learning", "abstract": "Model-agnostic meta-learning (MAML) is a meta-learning technique to train a model on a multitude of learning tasks in a way that primes the model for few-shot learning of new tasks. The MAML algorithm performs well on few-shot learning problems in classification, regression, and fine-tuning of policy gradients in reinforcement learning, but comes with the need for costly hyperparameter tuning for training stability. We address this shortcoming by introducing an extension to MAML, called Alpha MAML, to incorporate an online hyperparameter adaptation scheme that eliminates the need to tune meta-learning and learning rates. Our results with the Omniglot database demonstrate a substantial reduction in the need to tune MAML training hyperparameters and improvement to training stability with less sensitivity to hyperparameter choice.", "authors": ["Harkirat Singh Behl", "Atılım Güneş Baydin", "Philip H. S. Torr"], "categories": ["cs.LG", "cs.AI", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-05-17", "url": "https://arxiv.org/abs/1905.07435", "pdf_url": "https://arxiv.org/pdf/1905.07435v1", "arxiv_id": "1905.07435", "doi": null, "citation_count": 72, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4658}
{"id": "082cba3250274d6083c0bb86e58265294263f8f0b4c823ff38c1fb10cc27471a", "sources": ["arxiv", "semantic_scholar"], "title": "TapNet: Neural Network Augmented with Task-Adaptive Projection for Few-Shot Learning", "abstract": "Handling previously unseen tasks after given only a few training examples continues to be a tough challenge in machine learning. We propose TapNets, neural networks augmented with task-adaptive projection for improved few-shot learning. Here, employing a meta-learning strategy with episode-based training, a network and a set of per-class reference vectors are learned across widely varying tasks. At the same time, for every episode, features in the embedding space are linearly projected into a new space as a form of quick task-specific conditioning. The training loss is obtained based on a distance metric between the query and the reference vectors in the projection space. Excellent generalization results in this way. When tested on the Omniglot, miniImageNet and tieredImageNet datasets, we obtain state of the art classification accuracies under various few-shot scenarios.", "authors": ["Sung Whan Yoon", "Jun Seo", "Jaekyun Moon"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-05-16", "url": "https://arxiv.org/abs/1905.06549", "pdf_url": "https://arxiv.org/pdf/1905.06549v2", "arxiv_id": "1905.06549", "doi": null, "citation_count": 301, "influential_citation_count": 18, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.6394}
{"id": "5c02d84e68caf25f598335610edc7aa2fcc39b062657f50763cd58e53df026e0", "sources": ["arxiv", "semantic_scholar"], "title": "A Neural Network-Evolutionary Computational Framework for Remaining Useful Life Estimation of Mechanical Systems", "abstract": "This paper presents a framework for estimating the remaining useful life (RUL) of mechanical systems. The framework consists of a multi-layer perceptron and an evolutionary algorithm for optimizing the data-related parameters. The framework makes use of a strided time window to estimate the RUL for mechanical components. Tuning the data-related parameters can become a very time consuming task. The framework presented here automatically reshapes the data such that the efficiency of the model is increased. Furthermore, the complexity of the model is kept low, e.g. neural networks with few hidden layers and few neurons at each layer. Having simple models has several advantages like short training times and the capacity of being in environments with limited computational resources such as embedded systems. The proposed method is evaluated on the publicly available C-MAPSS dataset, its accuracy is compared against other state-of-the art methods for the same dataset.", "authors": ["David Laredo", "Zhaoyin Chen", "Oliver Schütze", "Jian-Qiao Sun"], "categories": ["cs.LG", "cs.NE", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics", "Medicine"], "published_date": "2019-05-15", "url": "https://arxiv.org/abs/1905.05918", "pdf_url": "https://arxiv.org/pdf/1905.05918v1", "arxiv_id": "1905.05918", "doi": "10.1016/j.neunet.2019.04.016", "citation_count": 54, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Neural Networks", "quality_score": 0.4351}
{"id": "14812c009f5d54c4f935f395d9d7ee0328a0b521dc122e25f4616d78aa56e296", "sources": ["arxiv", "semantic_scholar"], "title": "Embeddings and Representation Learning for Structured Data", "abstract": "Performing machine learning on structured data is complicated by the fact that such data does not have vectorial form. Therefore, multiple approaches have emerged to construct vectorial representations of structured data, from kernel and distance approaches to recurrent, recursive, and convolutional neural networks. Recent years have seen heightened attention in this demanding field of research and several new approaches have emerged, such as metric learning on structured data, graph convolutional neural networks, and recurrent decoder networks for structured data. In this contribution, we provide an high-level overview of the state-of-the-art in representation learning and embeddings for structured data across a wide range of machine learning fields.", "authors": ["Benjamin Paaßen", "Claudio Gallicchio", "Alessio Micheli", "Alessandro Sperduti"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-05-15", "url": "https://arxiv.org/abs/1905.06147", "pdf_url": "https://arxiv.org/pdf/1905.06147v1", "arxiv_id": "1905.06147", "doi": null, "citation_count": 8, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "The European Symposium on Artificial Neural Networks", "quality_score": 0.2386}
{"id": "579d2c3beafcec2368f9388660d630eb629837e1ec23db6fc887bb61d6a892ef", "sources": ["arxiv", "semantic_scholar"], "title": "Learning Generative Models across Incomparable Spaces", "abstract": "Generative Adversarial Networks have shown remarkable success in learning a distribution that faithfully recovers a reference distribution in its entirety. However, in some cases, we may want to only learn some aspects (e.g., cluster or manifold structure), while modifying others (e.g., style, orientation or dimension). In this work, we propose an approach to learn generative models across such incomparable spaces, and demonstrate how to steer the learned distribution towards target properties. A key component of our model is the Gromov-Wasserstein distance, a notion of discrepancy that compares distributions relationally rather than absolutely. While this framework subsumes current generative models in identically reproducing distributions, its inherent flexibility allows application to tasks in manifold learning, relational learning and cross-domain learning.", "authors": ["Charlotte Bunne", "David Alvarez-Melis", "Andreas Krause", "Stefanie Jegelka"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-05-14", "url": "https://arxiv.org/abs/1905.05461", "pdf_url": "https://arxiv.org/pdf/1905.05461v2", "arxiv_id": "1905.05461", "doi": "10.3929/ETHZ-B-000382654", "citation_count": 119, "influential_citation_count": 7, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.5198}
{"id": "5ff1a020dabc8c4b4303a2940c0804a72ae739dfed1099bd9fde0cef68b75c12", "sources": ["arxiv", "semantic_scholar"], "title": "Fast and Reliable Architecture Selection for Convolutional Neural Networks", "abstract": "The performance of a Convolutional Neural Network (CNN) depends on its hyperparameters, like the number of layers, kernel sizes, or the learning rate for example. Especially in smaller networks and applications with limited computational resources, optimisation is key. We present a fast and efficient approach for CNN architecture selection. Taking into account time consumption, precision and robustness, we develop a heuristic to quickly and reliably assess a network's performance. In combination with Bayesian optimisation (BO), to effectively cover the vast parameter space, our contribution offers a plain and powerful architecture search for this machine learning technique.", "authors": ["Lukas Hahn", "Lutz Roese-Koerner", "Klaus Friedrichs", "Anton Kummert"], "categories": ["cs.CV", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2019-05-06", "url": "https://arxiv.org/abs/1905.01924", "pdf_url": "https://arxiv.org/pdf/1905.01924v1", "arxiv_id": "1905.01924", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "The European Symposium on Artificial Neural Networks", "quality_score": 0.0}
{"id": "595392074956175759cd3d497d5f1e86e0c41b4d1c169e9040d8d40b2179ff2d", "sources": ["arxiv", "semantic_scholar"], "title": "Unsupervised Representation Learning with Minimax Distance Measures", "abstract": "We investigate the use of Minimax distances to extract in a nonparametric way the features that capture the unknown underlying patterns and structures in the data. We develop a general-purpose and computationally efficient framework to employ Minimax distances with many machine learning methods that perform on numerical data. We study both computing the pairwise Minimax distances for all pairs of objects and as well as computing the Minimax distances of all the objects to/from a fixed (test) object. We first efficiently compute the pairwise Minimax distances between the objects, using the equivalence of Minimax distances over a graph and over a minimum spanning tree constructed on that. Then, we perform an embedding of the pairwise Minimax distances into a new vector space, such that their squared Euclidean distances in the new space equal to the pairwise Minimax distances in the original space. We also study the case of having multiple pairwise Minimax matrices, instead of a single one. Thereby, we propose an embedding via first summing up the centered matrices and then performing an eigenvalue decomposition to obtain the relevant features. In the following, we study computing Minimax distances from a fixed (test) object which can be used for instance in K-nearest neighbor search. Similar to the case of all-pair pairwise Minimax distances, we develop an efficient and general-purpose algorithm that is applicable with any arbitrary base distance measure. Moreover, we investigate in detail the edges selected by the Minimax distances and thereby explore the ability of Minimax distances in detecting outlier objects. Finally, for each setting, we perform several experiments to demonstrate the effectiveness of our framework.", "authors": ["Morteza Haghir Chehreghani"], "categories": ["cs.LG", "cs.AI", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-04-27", "url": "https://arxiv.org/abs/1904.13223", "pdf_url": "https://arxiv.org/pdf/1904.13223v3", "arxiv_id": "1904.13223", "doi": "10.1007/s10994-020-05886-4", "citation_count": 14, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "Machine-mediated learning", "quality_score": 0.294}
{"id": "3e9bd46f4be18aa89a38f0dd89329b85639be35f39d44d4b8e5f44951a77c110", "sources": ["arxiv", "semantic_scholar"], "title": "Facilitating Bayesian Continual Learning by Natural Gradients and Stein Gradients", "abstract": "Continual learning aims to enable machine learning models to learn a general solution space for past and future tasks in a sequential manner. Conventional models tend to forget the knowledge of previous tasks while learning a new task, a phenomenon known as catastrophic forgetting. When using Bayesian models in continual learning, knowledge from previous tasks can be retained in two ways: 1). posterior distributions over the parameters, containing the knowledge gained from inference in previous tasks, which then serve as the priors for the following task; 2). coresets, containing knowledge of data distributions of previous tasks. Here, we show that Bayesian continual learning can be facilitated in terms of these two means through the use of natural gradients and Stein gradients respectively.", "authors": ["Yu Chen", "Tom Diethe", "Neil Lawrence"], "categories": ["cs.LG", "cs.AI", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-04-24", "url": "https://arxiv.org/abs/1904.10644", "pdf_url": "https://arxiv.org/pdf/1904.10644v1", "arxiv_id": "1904.10644", "doi": null, "citation_count": 15, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.301}
{"id": "1cb9a9ce979b3482fee857ad5402ced1877e1ce99d5667454038ff3c0e4bbddb", "sources": ["arxiv", "semantic_scholar"], "title": "Continual Learning with Self-Organizing Maps", "abstract": "Despite remarkable successes achieved by modern neural networks in a wide range of applications, these networks perform best in domain-specific stationary environments where they are trained only once on large-scale controlled data repositories. When exposed to non-stationary learning environments, current neural networks tend to forget what they had previously learned, a phenomena known as catastrophic forgetting. Most previous approaches to this problem rely on memory replay buffers which store samples from previously learned tasks, and use them to regularize the learning on new ones. This approach suffers from the important disadvantage of not scaling well to real-life problems in which the memory requirements become enormous. We propose a memoryless method that combines standard supervised neural networks with self-organizing maps to solve the continual learning problem. The role of the self-organizing map is to adaptively cluster the inputs into appropriate task contexts - without explicit labels - and allocate network resources accordingly. Thus, it selectively routes the inputs in accord with previous experience, ensuring that past learning is maintained and does not interfere with current learning. Out method is intuitive, memoryless, and performs on par with current state-of-the-art approaches on standard benchmarks.", "authors": ["Pouya Bashivan", "Martin Schrimpf", "Robert Ajemian", "Irina Rish", "Matthew Riemer", "Yuhai Tu"], "categories": ["cs.NE"], "fields_of_study": ["Computer Science"], "published_date": "2019-04-19", "url": "https://arxiv.org/abs/1904.09330", "pdf_url": "https://arxiv.org/pdf/1904.09330v1", "arxiv_id": "1904.09330", "doi": null, "citation_count": 17, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3138}
{"id": "fdc836043ed8ecd0768fd4db515dff29c2f66cf329b020cdefc67e5dafc9cb87", "sources": ["arxiv", "semantic_scholar"], "title": "Three scenarios for continual learning", "abstract": "Standard artificial neural networks suffer from the well-known issue of catastrophic forgetting, making continual or lifelong learning difficult for machine learning. In recent years, numerous methods have been proposed for continual learning, but due to differences in evaluation protocols it is difficult to directly compare their performance. To enable more structured comparisons, we describe three continual learning scenarios based on whether at test time task identity is provided and--in case it is not--whether it must be inferred. Any sequence of well-defined tasks can be performed according to each scenario. Using the split and permuted MNIST task protocols, for each scenario we carry out an extensive comparison of recently proposed continual learning methods. We demonstrate substantial differences between the three scenarios in terms of difficulty and in terms of how efficient different methods are. In particular, when task identity must be inferred (i.e., class incremental learning), we find that regularization-based approaches (e.g., elastic weight consolidation) fail and that replaying representations of previous experiences seems required for solving this scenario.", "authors": ["Gido M. van de Ven", "Andreas S. Tolias"], "categories": ["cs.LG", "cs.AI", "cs.CV", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-04-15", "url": "https://arxiv.org/abs/1904.07734", "pdf_url": "https://arxiv.org/pdf/1904.07734v1", "arxiv_id": "1904.07734", "doi": null, "citation_count": 1057, "influential_citation_count": 79, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.9515}
{"id": "48a2e72678690b407ffe680449b22b8333b8511c044200551c9614e7e668679d", "sources": ["arxiv", "semantic_scholar"], "title": "Transfer Learning with Sparse Associative Memories", "abstract": "In this paper, we introduce a novel layer designed to be used as the output of pre-trained neural networks in the context of classification. Based on Associative Memories, this layer can help design Deep Neural Networks which support incremental learning and that can be (partially) trained in real time on embedded devices. Experiments on the ImageNet dataset and other different domain specific datasets show that it is possible to design more flexible and faster-to-train Neural Networks at the cost of a slight decrease in accuracy.", "authors": ["Quentin Jodelet", "Vincent Gripon", "Masafumi Hagiwara"], "categories": ["cs.LG", "cs.CV", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-04-04", "url": "https://arxiv.org/abs/1904.02420", "pdf_url": "https://arxiv.org/pdf/1904.02420v3", "arxiv_id": "1904.02420", "doi": "10.1007/978-3-030-30487-4_39", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "International Conference on Artificial Neural Networks", "quality_score": 0.0}
{"id": "6985f1e254553076b92eebf33f8ad47453098cb8957d14b7c32ca7a22f22311a", "sources": ["arxiv", "semantic_scholar"], "title": "Learn to Grow: A Continual Structure Learning Framework for Overcoming Catastrophic Forgetting", "abstract": "Addressing catastrophic forgetting is one of the key challenges in continual learning where machine learning systems are trained with sequential or streaming tasks. Despite recent remarkable progress in state-of-the-art deep learning, deep neural networks (DNNs) are still plagued with the catastrophic forgetting problem. This paper presents a conceptually simple yet general and effective framework for handling catastrophic forgetting in continual learning with DNNs. The proposed method consists of two components: a neural structure optimization component and a parameter learning and/or fine-tuning component. By separating the explicit neural structure learning and the parameter estimation, not only is the proposed method capable of evolving neural structures in an intuitively meaningful way, but also shows strong capabilities of alleviating catastrophic forgetting in experiments. Furthermore, the proposed method outperforms all other baselines on the permuted MNIST dataset, the split CIFAR100 dataset and the Visual Domain Decathlon dataset in continual learning setting.", "authors": ["Xilai Li", "Yingbo Zhou", "Tianfu Wu", "Richard Socher", "Caiming Xiong"], "categories": ["cs.LG", "cs.CV"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-03-31", "url": "https://arxiv.org/abs/1904.00310", "pdf_url": "https://arxiv.org/pdf/1904.00310v3", "arxiv_id": "1904.00310", "doi": null, "citation_count": 525, "influential_citation_count": 16, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.6802}
{"id": "cc8e1bed4d2dd901fcf87784e8c17dd87adbb0ce2cc1a4c302ba2e74c4085c19", "sources": ["arxiv", "semantic_scholar"], "title": "On-line learning dynamics of ReLU neural networks using statistical physics techniques", "abstract": "We introduce exact macroscopic on-line learning dynamics of two-layer neural networks with ReLU units in the form of a system of differential equations, using techniques borrowed from statistical physics. For the first experiments, numerical solutions reveal similar behavior compared to sigmoidal activation researched in earlier work. In these experiments the theoretical results show good correspondence with simulations. In ove-rrealizable and unrealizable learning scenarios, the learning behavior of ReLU networks shows distinctive characteristics compared to sigmoidal networks.", "authors": ["Michiel Straat", "Michael Biehl"], "categories": ["cs.LG", "cond-mat.dis-nn", "stat.ML"], "fields_of_study": ["Computer Science", "Physics", "Mathematics"], "published_date": "2019-03-18", "url": "https://arxiv.org/abs/1903.07378", "pdf_url": "https://arxiv.org/pdf/1903.07378v1", "arxiv_id": "1903.07378", "doi": null, "citation_count": 11, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "The European Symposium on Artificial Neural Networks", "quality_score": 0.2698}
{"id": "2a28853e157a7139e2a487d0006fd719f3e513b09b2e5d52936e6c4443987e59", "sources": ["arxiv", "semantic_scholar"], "title": "Communication-Efficient Federated Deep Learning with Asynchronous Model Update and Temporally Weighted Aggregation", "abstract": "Federated learning obtains a central model on the server by aggregating models trained locally on clients. As a result, federated learning does not require clients to upload their data to the server, thereby preserving the data privacy of the clients. One challenge in federated learning is to reduce the client-server communication since the end devices typically have very limited communication bandwidth. This paper presents an enhanced federated learning technique by proposing a synchronous learning strategy on the clients and a temporally weighted aggregation of the local models on the server. In the asynchronous learning strategy, different layers of the deep neural networks are categorized into shallow and deeps layers and the parameters of the deep layers are updated less frequently than those of the shallow layers. Furthermore, a temporally weighted aggregation strategy is introduced on the server to make use of the previously trained local models, thereby enhancing the accuracy and convergence of the central model. The proposed algorithm is empirically on two datasets with different deep neural networks. Our results demonstrate that the proposed asynchronous federated deep learning outperforms the baseline algorithm both in terms of communication cost and model accuracy.", "authors": ["Yang Chen", "Xiaoyan Sun", "Yaochu Jin"], "categories": ["cs.LG", "cs.AI", "cs.DC", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics", "Medicine"], "published_date": "2019-03-18", "url": "https://arxiv.org/abs/1903.07424", "pdf_url": "https://arxiv.org/pdf/1903.07424v1", "arxiv_id": "1903.07424", "doi": "10.1109/TNNLS.2019.2953131", "citation_count": 527, "influential_citation_count": 28, "has_code": false, "code_url": null, "venue": "IEEE Transactions on Neural Networks and Learning Systems", "quality_score": 0.7312}
{"id": "8c42a0cebcdd399fed324369a7fc5d72d7bc69c26140e624c375caff4abca227", "sources": ["arxiv", "semantic_scholar"], "title": "Continual Learning in Practice", "abstract": "This paper describes a reference architecture for self-maintaining systems that can learn continually, as data arrives. In environments where data evolves, we need architectures that manage Machine Learning (ML) models in production, adapt to shifting data distributions, cope with outliers, retrain when necessary, and adapt to new tasks. This represents continual AutoML or Automatically Adaptive Machine Learning. We describe the challenges and proposes a reference architecture.", "authors": ["Tom Diethe", "Tom Borchert", "Eno Thereska", "Borja Balle", "Neil Lawrence"], "categories": ["stat.ML", "cs.LG"], "fields_of_study": ["Mathematics", "Computer Science"], "published_date": "2019-03-12", "url": "https://arxiv.org/abs/1903.05202", "pdf_url": "https://arxiv.org/pdf/1903.05202v2", "arxiv_id": "1903.05202", "doi": null, "citation_count": 78, "influential_citation_count": 5, "has_code": false, "code_url": null, "venue": "Neural Information Processing Systems", "quality_score": 0.4744}
{"id": "a9c86f17ad94c38deb380c073f1bb6afc2b7a99ece866a4659cd98b6f23816a4", "sources": ["arxiv", "semantic_scholar"], "title": "Continual Learning via Neural Pruning", "abstract": "We introduce Continual Learning via Neural Pruning (CLNP), a new method aimed at lifelong learning in fixed capacity models based on neuronal model sparsification. In this method, subsequent tasks are trained using the inactive neurons and filters of the sparsified network and cause zero deterioration to the performance of previous tasks. In order to deal with the possible compromise between model sparsity and performance, we formalize and incorporate the concept of graceful forgetting: the idea that it is preferable to suffer a small amount of forgetting in a controlled manner if it helps regain network capacity and prevents uncontrolled loss of performance during the training of future tasks. CLNP also provides simple continual learning diagnostic tools in terms of the number of free neurons left for the training of future tasks as well as the number of neurons that are being reused. In particular, we see in experiments that CLNP verifies and automatically takes advantage of the fact that the features of earlier layers are more transferable. We show empirically that CLNP leads to significantly improved results over current weight elasticity based methods.", "authors": ["Siavash Golkar", "Michael Kagan", "Kyunghyun Cho"], "categories": ["cs.LG", "cs.NE", "q-bio.NC", "stat.ML"], "fields_of_study": ["Computer Science", "Biology", "Mathematics"], "published_date": "2019-03-11", "url": "https://arxiv.org/abs/1903.04476", "pdf_url": "https://arxiv.org/pdf/1903.04476v1", "arxiv_id": "1903.04476", "doi": null, "citation_count": 181, "influential_citation_count": 10, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.565}
{"id": "1a9a3535c46cebee31e28ad7fd39f9761baa271e8c45975afac5df4d99f68a7c", "sources": ["arxiv", "semantic_scholar"], "title": "Complementary Learning for Overcoming Catastrophic Forgetting Using Experience Replay", "abstract": "Despite huge success, deep networks are unable to learn effectively in sequential multitask learning settings as they forget the past learned tasks after learning new tasks. Inspired from complementary learning systems theory, we address this challenge by learning a generative model that couples the current task to the past learned tasks through a discriminative embedding space. We learn an abstract level generative distribution in the embedding that allows the generation of data points to represent the experience. We sample from this distribution and utilize experience replay to avoid forgetting and simultaneously accumulate new knowledge to the abstract distribution in order to couple the current task with past experience. We demonstrate theoretically and empirically that our framework learns a distribution in the embedding that is shared across all task and as a result tackles catastrophic forgetting.", "authors": ["Mohammad Rostami", "Soheil Kolouri", "Praveen K. Pilly"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-03-11", "url": "https://arxiv.org/abs/1903.04566", "pdf_url": "https://arxiv.org/pdf/1903.04566v2", "arxiv_id": "1903.04566", "doi": "10.24963/ijcai.2019/463", "citation_count": 82, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "International Joint Conference on Artificial Intelligence", "quality_score": 0.4798}
{"id": "69305eb94930e06b6f7fb0b02c6706b4a82086da4a2ba2061ee43568837d87a2", "sources": ["arxiv", "semantic_scholar"], "title": "Interpolation Consistency Training for Semi-Supervised Learning", "abstract": "We introduce Interpolation Consistency Training (ICT), a simple and computation efficient algorithm for training Deep Neural Networks in the semi-supervised learning paradigm. ICT encourages the prediction at an interpolation of unlabeled points to be consistent with the interpolation of the predictions at those points. In classification problems, ICT moves the decision boundary to low-density regions of the data distribution. Our experiments show that ICT achieves state-of-the-art performance when applied to standard neural network architectures on the CIFAR-10 and SVHN benchmark datasets. Our theoretical analysis shows that ICT corresponds to a certain type of data-adaptive regularization with unlabeled points which reduces overfitting to labeled points under high confidence values.", "authors": ["Vikas Verma", "Kenji Kawaguchi", "Alex Lamb", "Juho Kannala", "Arno Solin", "Yoshua Bengio", "David Lopez-Paz"], "categories": ["stat.ML", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science", "Medicine", "Mathematics"], "published_date": "2019-03-09", "url": "https://arxiv.org/abs/1903.03825", "pdf_url": "https://arxiv.org/pdf/1903.03825v5", "arxiv_id": "1903.03825", "doi": "10.1016/j.neunet.2021.10.008", "citation_count": 911, "influential_citation_count": 105, "has_code": false, "code_url": null, "venue": "International Joint Conference on Artificial Intelligence", "quality_score": 1.0}
{"id": "5f3b16e3874ed26bd53f48886b51b43c8b56084760f4dd5d057d031496f24a2e", "sources": ["arxiv", "semantic_scholar"], "title": "Transfer Learning Using Ensemble Neural Networks for Organic Solar Cell Screening", "abstract": "Organic Solar Cells are a promising technology for solving the clean energy crisis in the world. However, generating candidate chemical compounds for solar cells is a time-consuming process requiring thousands of hours of laboratory analysis. For a solar cell, the most important property is the power conversion efficiency which is dependent on the highest occupied molecular orbitals (HOMO) values of the donor molecules. Recently, machine learning techniques have proved to be very useful in building predictive models for HOMO values of donor structures of Organic Photovoltaic Cells (OPVs). Since experimental datasets are limited in size, current machine learning models are trained on data derived from calculations based on density functional theory (DFT). Molecular line notations such as SMILES or InChI are popular input representations for describing the molecular structure of donor molecules. The two types of line representations encode different information, such as SMILES defines the bond types while InChi defines protonation. In this work, we present an ensemble deep neural network architecture, called SINet, which harnesses both the SMILES and InChI molecular representations to predict HOMO values and leverage the potential of transfer learning from a sizeable DFT-computed dataset- Harvard CEP to build more robust predictive models for relatively smaller HOPV datasets. Harvard CEP dataset contains molecular structures and properties for 2.3 million candidate donor structures for OPV while HOPV contains DFT-computed and experimental values of 350 and 243 molecules respectively. Our results demonstrate significant performance improvement from the use of transfer learning and leveraging both molecular representations.", "authors": ["Arindam Paul", "Dipendra Jha", "Reda Al-Bahrani", "Wei-keng Liao", "Alok Choudhary", "Ankit Agrawal"], "categories": ["cs.LG", "physics.chem-ph", "stat.ML"], "fields_of_study": ["Computer Science", "Physics", "Mathematics"], "published_date": "2019-03-07", "url": "https://arxiv.org/abs/1903.03178", "pdf_url": "https://arxiv.org/pdf/1903.03178v4", "arxiv_id": "1903.03178", "doi": "10.1109/IJCNN.2019.8852446", "citation_count": 22, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "IEEE International Joint Conference on Neural Network", "quality_score": 0.3404}
{"id": "4764c15a5844e41937eec236a617e3bd4af382fe4860672b36e4c6cb0598e85a", "sources": ["arxiv", "semantic_scholar"], "title": "PDP: A General Neural Framework for Learning Constraint Satisfaction Solvers", "abstract": "There have been recent efforts for incorporating Graph Neural Network models for learning full-stack solvers for constraint satisfaction problems (CSP) and particularly Boolean satisfiability (SAT). Despite the unique representational power of these neural embedding models, it is not clear how the search strategy in the learned models actually works. On the other hand, by fixing the search strategy (e.g. greedy search), we would effectively deprive the neural models of learning better strategies than those given. In this paper, we propose a generic neural framework for learning CSP solvers that can be described in terms of probabilistic inference and yet learn search strategies beyond greedy search. Our framework is based on the idea of propagation, decimation and prediction (and hence the name PDP) in graphical models, and can be trained directly toward solving CSP in a fully unsupervised manner via energy minimization, as shown in the paper. Our experimental results demonstrate the effectiveness of our framework for SAT solving compared to both neural and the state-of-the-art baselines.", "authors": ["Saeed Amizadeh", "Sergiy Matusevych", "Markus Weimer"], "categories": ["cs.LG", "cs.LO", "cs.NE", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-03-05", "url": "https://arxiv.org/abs/1903.01969", "pdf_url": "https://arxiv.org/pdf/1903.01969v1", "arxiv_id": "1903.01969", "doi": null, "citation_count": 23, "influential_citation_count": 3, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3451}
{"id": "c52d6cc088146c56d993eafc17c4b2d87dfd3123c121493079e8eb99c153a1ac", "sources": ["arxiv", "semantic_scholar"], "title": "Scalable and Order-robust Continual Learning with Additive Parameter Decomposition", "abstract": "While recent continual learning methods largely alleviate the catastrophic problem on toy-sized datasets, some issues remain to be tackled to apply them to real-world problem domains. First, a continual learning model should effectively handle catastrophic forgetting and be efficient to train even with a large number of tasks. Secondly, it needs to tackle the problem of order-sensitivity, where the performance of the tasks largely varies based on the order of the task arrival sequence, as it may cause serious problems where fairness plays a critical role (e.g. medical diagnosis). To tackle these practical challenges, we propose a novel continual learning method that is scalable as well as order-robust, which instead of learning a completely shared set of weights, represents the parameters for each task as a sum of task-shared and sparse task-adaptive parameters. With our Additive Parameter Decomposition (APD), the task-adaptive parameters for earlier tasks remain mostly unaffected, where we update them only to reflect the changes made to the task-shared parameters. This decomposition of parameters effectively prevents catastrophic forgetting and order-sensitivity, while being computation- and memory-efficient. Further, we can achieve even better scalability with APD using hierarchical knowledge consolidation, which clusters the task-adaptive parameters to obtain hierarchically shared parameters. We validate our network with APD, APD-Net, on multiple benchmark datasets against state-of-the-art continual learning methods, which it largely outperforms in accuracy, scalability, and order-robustness.", "authors": ["Jaehong Yoon", "Saehoon Kim", "Eunho Yang", "Sung Ju Hwang"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-02-25", "url": "https://arxiv.org/abs/1902.09432", "pdf_url": "https://arxiv.org/pdf/1902.09432v3", "arxiv_id": "1902.09432", "doi": null, "citation_count": 209, "influential_citation_count": 20, "has_code": false, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.6611}
{"id": "e70ab08b1fa983c335ffd2a37a9f8eb013e74096d323fbcd32088efdd9359ac3", "sources": ["arxiv", "semantic_scholar"], "title": "Deep Bayesian Multi-Target Learning for Recommender Systems", "abstract": "With the increasing variety of services that e-commerce platforms provide, criteria for evaluating their success become also increasingly multi-targeting. This work introduces a multi-target optimization framework with Bayesian modeling of the target events, called Deep Bayesian Multi-Target Learning (DBMTL). In this framework, target events are modeled as forming a Bayesian network, in which directed links are parameterized by hidden layers, and learned from training samples. The structure of Bayesian network is determined by model selection. We applied the framework to Taobao live-streaming recommendation, to simultaneously optimize (and strike a balance) on targets including click-through rate, user stay time in live room, purchasing behaviors and interactions. Significant improvement has been observed for the proposed method over other MTL frameworks and the non-MTL model. Our practice shows that with an integrated causality structure, we can effectively make the learning of a target benefit from other targets, creating significant synergy effects that improve all targets. The neural network construction guided by DBMTL fits in with the general probabilistic model connecting features and multiple targets, taking weaker assumption than the other methods discussed in this paper. This theoretical generality brings about practical generalization power over various targets distributions, including sparse targets and continuous-value ones.", "authors": ["Qi Wang", "Zhihui Ji", "Huasheng Liu", "Binqiang Zhao"], "categories": ["cs.LG", "cs.IR", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-02-25", "url": "https://arxiv.org/abs/1902.09154", "pdf_url": "https://arxiv.org/pdf/1902.09154v1", "arxiv_id": "1902.09154", "doi": null, "citation_count": 14, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.294}
{"id": "87e6f5f778c6c9d5b97c59bcd76177dec6f12328e87f630311628dda440fe4f2", "sources": ["arxiv", "semantic_scholar"], "title": "Differentially Private Continual Learning", "abstract": "Catastrophic forgetting can be a significant problem for institutions that must delete historic data for privacy reasons. For example, hospitals might not be able to retain patient data permanently. But neural networks trained on recent data alone will tend to forget lessons learned on old data. We present a differentially private continual learning framework based on variational inference. We estimate the likelihood of past data given the current model using differentially private generative models of old datasets.", "authors": ["Sebastian Farquhar", "Yarin Gal"], "categories": ["stat.ML", "cs.LG"], "fields_of_study": ["Mathematics", "Computer Science"], "published_date": "2019-02-18", "url": "https://arxiv.org/abs/1902.06497", "pdf_url": "https://arxiv.org/pdf/1902.06497v1", "arxiv_id": "1902.06497", "doi": null, "citation_count": 12, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2785}
{"id": "81fab21e758913394f5a79a92a7435887e4f080ef4663d3b624063d37e254589", "sources": ["arxiv", "semantic_scholar"], "title": "A Unifying Bayesian View of Continual Learning", "abstract": "Some machine learning applications require continual learning - where data comes in a sequence of datasets, each is used for training and then permanently discarded. From a Bayesian perspective, continual learning seems straightforward: Given the model posterior one would simply use this as the prior for the next task. However, exact posterior evaluation is intractable with many models, especially with Bayesian neural networks (BNNs). Instead, posterior approximations are often sought. Unfortunately, when posterior approximations are used, prior-focused approaches do not succeed in evaluations designed to capture properties of realistic continual learning use cases. As an alternative to prior-focused methods, we introduce a new approximate Bayesian derivation of the continual learning loss. Our loss does not rely on the posterior from earlier tasks, and instead adapts the model itself by changing the likelihood term. We call these approaches likelihood-focused. We then combine prior- and likelihood-focused methods into one objective, tying the two views together under a single unifying framework of approximate Bayesian continual learning.", "authors": ["Sebastian Farquhar", "Yarin Gal"], "categories": ["stat.ML", "cs.LG"], "fields_of_study": ["Mathematics", "Computer Science"], "published_date": "2019-02-18", "url": "https://arxiv.org/abs/1902.06494", "pdf_url": "https://arxiv.org/pdf/1902.06494v1", "arxiv_id": "1902.06494", "doi": null, "citation_count": 80, "influential_citation_count": 7, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4771}
{"id": "d904892f020e2e45c11f27e6cb00fb01337d6a95791c834a4be5a2df37ecf5c6", "sources": ["arxiv", "semantic_scholar"], "title": "Scaling Limits of Wide Neural Networks with Weight Sharing: Gaussian Process Behavior, Gradient Independence, and Neural Tangent Kernel Derivation", "abstract": "Several recent trends in machine learning theory and practice, from the design of state-of-the-art Gaussian Process to the convergence analysis of deep neural nets (DNNs) under stochastic gradient descent (SGD), have found it fruitful to study wide random neural networks. Central to these approaches are certain scaling limits of such networks. We unify these results by introducing a notion of a straightline \\emph{tensor program} that can express most neural network computations, and we characterize its scaling limit when its tensors are large and randomized. From our framework follows (1) the convergence of random neural networks to Gaussian processes for architectures such as recurrent neural networks, convolutional neural networks, residual networks, attention, and any combination thereof, with or without batch normalization; (2) conditions under which the \\emph{gradient independence assumption} -- that weights in backpropagation can be assumed to be independent from weights in the forward pass -- leads to correct computation of gradient dynamics, and corrections when it does not; (3) the convergence of the Neural Tangent Kernel, a recently proposed kernel used to predict training dynamics of neural networks under gradient descent, at initialization for all architectures in (1) without batch normalization. Mathematically, our framework is general enough to rederive classical random matrix results such as the semicircle and the Marchenko-Pastur laws, as well as recent results in neural network Jacobian singular values. We hope our work opens a way toward design of even stronger Gaussian Processes, initialization schemes to avoid gradient explosion/vanishing, and deeper understanding of SGD dynamics in modern architectures.", "authors": ["Greg Yang"], "categories": ["cs.NE", "cond-mat.dis-nn", "cs.LG", "math-ph", "stat.ML"], "fields_of_study": ["Computer Science", "Physics", "Mathematics"], "published_date": "2019-02-13", "url": "https://arxiv.org/abs/1902.04760", "pdf_url": "https://arxiv.org/pdf/1902.04760v3", "arxiv_id": "1902.04760", "doi": null, "citation_count": 311, "influential_citation_count": 36, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.7841}
{"id": "106f6a9ebe3ffdedaa5d5af5565185ad0f20edbda8549eadff0a1cf8d91aa1e8", "sources": ["arxiv", "semantic_scholar"], "title": "Controlled Forgetting: Targeted Stimulation and Dopaminergic Plasticity Modulation for Unsupervised Lifelong Learning in Spiking Neural Networks", "abstract": "Stochastic gradient descent requires that training samples be drawn from a uniformly random distribution of the data. For a deployed system that must learn online from an uncontrolled and unknown environment, the ordering of input samples often fails to meet this criterion, making lifelong learning a difficult challenge. We exploit the locality of the unsupervised Spike Timing Dependent Plasticity (STDP) learning rule to target local representations in a Spiking Neural Network (SNN) to adapt to novel information while protecting essential information in the remainder of the SNN from catastrophic forgetting. In our Controlled Forgetting Networks (CFNs), novel information triggers stimulated firing and heterogeneously modulated plasticity, inspired by biological dopamine signals, to cause rapid and isolated adaptation in the synapses of neurons associated with outlier information. This targeting controls the forgetting process in a way that reduces the degradation of accuracy for older tasks while learning new tasks. Our experimental results on the MNIST dataset validate the capability of CFNs to learn successfully over time from an unknown, changing environment, achieving 95.36% accuracy, which we believe is the best unsupervised accuracy ever achieved by a fixed-size, single-layer SNN on a completely disjoint MNIST dataset.", "authors": ["Jason M. Allred", "Kaushik Roy"], "categories": ["cs.NE", "cs.LG"], "fields_of_study": ["Computer Science", "Medicine"], "published_date": "2019-02-08", "url": "https://arxiv.org/abs/1902.03187", "pdf_url": "https://arxiv.org/pdf/1902.03187v2", "arxiv_id": "1902.03187", "doi": "10.3389/fnins.2020.00007", "citation_count": 38, "influential_citation_count": 4, "has_code": false, "code_url": null, "venue": "Frontiers in Neuroscience", "quality_score": 0.3978}
{"id": "79b5a8b6b873d83ad123b3eca4bf14a5e1d76a1f4543b819c5e863ff8d3cba62", "sources": ["arxiv", "semantic_scholar"], "title": "On ADMM in Deep Learning: Convergence and Saturation-Avoidance", "abstract": "In this paper, we develop an alternating direction method of multipliers (ADMM) for deep neural networks training with sigmoid-type activation functions (called \\textit{sigmoid-ADMM pair}), mainly motivated by the gradient-free nature of ADMM in avoiding the saturation of sigmoid-type activations and the advantages of deep neural networks with sigmoid-type activations (called deep sigmoid nets) over their rectified linear unit (ReLU) counterparts (called deep ReLU nets) in terms of approximation. In particular, we prove that the approximation capability of deep sigmoid nets is not worse than that of deep ReLU nets by showing that ReLU activation function can be well approximated by deep sigmoid nets with two hidden layers and finitely many free parameters but not vice-verse. We also establish the global convergence of the proposed ADMM for the nonlinearly constrained formulation of the deep sigmoid nets training from arbitrary initial points to a Karush-Kuhn-Tucker (KKT) point at a rate of order ${\\cal O}(1/k)$. Besides sigmoid activation, such a convergence theorem holds for a general class of smooth activations. Compared with the widely used stochastic gradient descent (SGD) algorithm for the deep ReLU nets training (called ReLU-SGD pair), the proposed sigmoid-ADMM pair is practically stable with respect to the algorithmic hyperparameters including the learning rate, initial schemes and the pro-processing of the input data. Moreover, we find that to approximate and learn simple but important functions the proposed sigmoid-ADMM pair numerically outperforms the ReLU-SGD pair.", "authors": ["Jinshan Zeng", "Shao-Bo Lin", "Yuan Yao", "Ding-Xuan Zhou"], "categories": ["cs.LG", "math.OC", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-02-06", "url": "https://arxiv.org/abs/1902.02060", "pdf_url": "https://arxiv.org/pdf/1902.02060v3", "arxiv_id": "1902.02060", "doi": null, "citation_count": 38, "influential_citation_count": 3, "has_code": false, "code_url": null, "venue": "Journal of machine learning research", "quality_score": 0.3978}
{"id": "92023f6ecc2141ec54bb624b3d940a31f0dcb2a83ccb4bcc8aeee8589c247f87", "sources": ["arxiv", "semantic_scholar"], "title": "Generalisation dynamics of online learning in over-parameterised neural networks", "abstract": "Deep neural networks achieve stellar generalisation on a variety of problems, despite often being large enough to easily fit all their training data. Here we study the generalisation dynamics of two-layer neural networks in a teacher-student setup, where one network, the student, is trained using stochastic gradient descent (SGD) on data generated by another network, called the teacher. We show how for this problem, the dynamics of SGD are captured by a set of differential equations. In particular, we demonstrate analytically that the generalisation error of the student increases linearly with the network size, with other relevant parameters held constant. Our results indicate that achieving good generalisation in neural networks depends on the interplay of at least the algorithm, its learning rate, the model architecture, and the data set.", "authors": ["Sebastian Goldt", "Madhu S. Advani", "Andrew M. Saxe", "Florent Krzakala", "Lenka Zdeborová"], "categories": ["stat.ML", "cond-mat.dis-nn", "cond-mat.stat-mech", "cs.LG"], "fields_of_study": ["Computer Science", "Mathematics", "Physics"], "published_date": "2019-01-25", "url": "https://arxiv.org/abs/1901.09085", "pdf_url": "https://arxiv.org/pdf/1901.09085v1", "arxiv_id": "1901.09085", "doi": null, "citation_count": 15, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.301}
{"id": "1ad24db00ba0caf229ef7fd0954100ba486e66f7bc4fbea36e105d579cadd032", "sources": ["arxiv", "semantic_scholar"], "title": "Unsupervised Learning of Neural Networks to Explain Neural Networks (extended abstract)", "abstract": "This paper presents an unsupervised method to learn a neural network, namely an explainer, to interpret a pre-trained convolutional neural network (CNN), i.e., the explainer uses interpretable visual concepts to explain features in middle conv-layers of a CNN. Given feature maps of a conv-layer of the CNN, the explainer performs like an auto-encoder, which decomposes the feature maps into object-part features. The object-part features are learned to reconstruct CNN features without much loss of information. We can consider the disentangled representations of object parts a paraphrase of CNN features, which help people understand the knowledge encoded by the CNN. More crucially, we learn the explainer via knowledge distillation without using any annotations of object parts or textures for supervision. In experiments, our method was widely used to interpret features of different benchmark CNNs, and explainers significantly boosted the feature interpretability without hurting the discrimination power of the CNNs.", "authors": ["Quanshi Zhang", "Yu Yang", "Ying Nian Wu"], "categories": ["cs.LG", "cs.AI", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2019-01-21", "url": "https://arxiv.org/abs/1901.07538", "pdf_url": "https://arxiv.org/pdf/1901.07538v1", "arxiv_id": "1901.07538", "doi": null, "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.0753}
{"id": "8c815760bbedba307ef530e4b189758ccd7995c56ecf61034fe7200add06d0f7", "sources": ["arxiv", "semantic_scholar"], "title": "Hierarchical Attentional Hybrid Neural Networks for Document Classification", "abstract": "Document classification is a challenging task with important applications. The deep learning approaches to the problem have gained much attention recently. Despite the progress, the proposed models do not incorporate the knowledge of the document structure in the architecture efficiently and not take into account the contexting importance of words and sentences. In this paper, we propose a new approach based on a combination of convolutional neural networks, gated recurrent units, and attention mechanisms for document classification tasks. The main contribution of this work is the use of convolution layers to extract more meaningful, generalizable and abstract features by the hierarchical representation. The proposed method in this paper improves the results of the current attention-based approaches for document classification.", "authors": ["Jader Abreu", "Luis Fred", "David Macêdo", "Cleber Zanchettin"], "categories": ["cs.CL", "cs.AI", "cs.LG", "cs.NE"], "fields_of_study": ["Computer Science"], "published_date": "2019-01-20", "url": "https://arxiv.org/abs/1901.06610", "pdf_url": "https://arxiv.org/pdf/1901.06610v2", "arxiv_id": "1901.06610", "doi": "10.1007/978-3-030-30493-5_39", "citation_count": 38, "influential_citation_count": 5, "has_code": false, "code_url": null, "venue": "International Conference on Artificial Neural Networks", "quality_score": 0.3978}