_id stringlengths 4 10 | text stringlengths 0 18.4k | title stringlengths 0 8.56k |
|---|---|---|
d249431362 | State-of-the-art text classification models are becoming increasingly reliant on deep neural networks (DNNs). Due to their black-box nature, faithful and robust explanation methods need to accompany classifiers for deployment in reallife scenarios. However, it has been shown in vision applications that explanation methods are susceptible to local, imperceptible perturbations that can significantly alter the explanations without changing the predicted classes. We show here that the existence of such perturbations extends to text classifiers as well. Specifically, we introduce TEXTEXPLANATIONFOOLER (TEF), a novel explanation attack algorithm that alters text input samples imperceptibly so that the outcome of widely-used explanation methods changes considerably while leaving classifier predictions unchanged. We evaluate the performance of the attribution robustness estimation performance in TEF on five sequence classification datasets, utilizing three DNN architectures and three transformer architectures for each dataset. TEF can significantly decrease the correlation between unchanged and perturbed input attributions, which shows that all models and explanation methods are susceptible to TEF perturbations. Moreover, we evaluate how the perturbations transfer to other model architectures and attribution methods, and show that TEF perturbations are also effective in scenarios where the target model and explanation method are unknown. Finally, we introduce a semi-universal attack that is able to compute fast, computationally light perturbations with no knowledge of the attacked classifier nor explanation method. Overall, our work shows that explanations in text classifiers are very fragile and users need to carefully address their robustness before relying on them in critical applications. | FOOLING EXPLANATIONS IN TEXT CLASSIFIERS |
d263831268 | In this paper, we investigate the behavior of gradient descent algorithms in physics-informed machine learning methods like PINNs, which minimize residuals connected to partial differential equations (PDEs).Our key result is that the difficulty in training these models is closely related to the conditioning of a specific differential operator.This operator, in turn, is associated to the Hermitian square of the differential operator of the underlying PDE.If this operator is ill-conditioned, it results in slow or infeasible training.Therefore, preconditioning this operator is crucial.We employ both rigorous mathematical analysis and empirical evaluations to investigate various strategies, explaining how they better condition this critical operator, and consequently improve training. | AN OPERATOR PRECONDITIONING PERSPECTIVE ON TRAINING IN PHYSICS-INFORMED MACHINE LEARNING |
d53100211 | Lack of performance when it comes to continual learning over non-stationary distributions of data remains a major challenge in scaling neural network learning to more human realistic settings. In this work we propose a new conceptualization of the continual learning problem in terms of a temporally symmetric trade-off between transfer and interference that can be optimized by enforcing gradient alignment across examples. We then propose a new algorithm, Meta-Experience Replay (MER), that directly exploits this view by combining experience replay with optimization based meta-learning. This method learns parameters that make interference based on future gradients less likely and transfer based on future gradients more likely. 1 We conduct experiments across continual lifelong supervised learning benchmarks and non-stationary reinforcement learning environments demonstrating that our approach consistently outperforms recently proposed baselines for continual learning. Our experiments show that the gap between the performance of MER and baseline algorithms grows both as the environment gets more non-stationary and as the fraction of the total experiences stored gets smaller.1 | LEARNING TO LEARN WITHOUT FORGETTING BY MAXIMIZING TRANSFER AND MINIMIZING INTERFERENCE |
d247411282 | Recently, post-training quantization (PTQ) has driven much attention to produce efficient neural networks without long-time retraining. Despite its low cost, current PTQ works tend to fail under the extremely low-bit setting. In this study, we pioneeringly confirm that properly incorporating activation quantization into the PTQ reconstruction benefits the final accuracy. To deeply understand the inherent reason, a theoretical framework is established, indicating that the flatness of the optimized low-bit model on calibration and test data is crucial. Based on the conclusion, a simple yet effective approach dubbed as QDROP is proposed, which randomly drops the quantization of activations during PTQ. Extensive experiments on various tasks including computer vision (image classification, object detection) and natural language processing (text classification and question answering) prove its superiority. With QDROP, the limit of PTQ is pushed to the 2-bit activation for the first time and the accuracy boost can be up to 51.49%. Without bells and whistles, QDROP establishes a new state of the art for PTQ. Our code is available at https://github.com/wimh966/QDrop and has been integrated into MQBench (https://github.com/ModelTC/MQBench). | QDROP: RANDOMLY DROPPING QUANTIZATION FOR EXTREMELY LOW-BIT POST-TRAINING QUANTIZATION |
d50783524 | Human annotation for syntactic parsing is expensive, and large resources are available only for a fraction of languages. A question we ask is whether one can leverage abundant unlabeled texts to improve syntactic parsers, beyond just using the texts to obtain more generalisable lexical features (i.e. beyond word embeddings). To this end, we propose a novel latent-variable generative model for semisupervised syntactic dependency parsing. As exact inference is intractable, we introduce a differentiable relaxation to obtain approximate samples and compute gradients with respect to the parser parameters. Our method (Differentiable Perturb-and-Parse) relies on differentiable dynamic programming over stochastically perturbed edge scores. We demonstrate effectiveness of our approach with experiments on English, French and Swedish. | Differentiable Perturb-and-Parse: Semi-Supervised Parsing with a Structured Variational Autoencoder |
d3538627 | Questions that require counting a variety of objects in images remain a major challenge in visual question answering (VQA). The most common approaches to VQA involve either classifying answers based on fixed length representations of both the image and question or summing fractional counts estimated from each section of the image. In contrast, we treat counting as a sequential decision process and force our model to make discrete choices of what to count. Specifically, the model sequentially selects from detected objects and learns interactions between objects that influence subsequent selections. A distinction of our approach is its intuitive and interpretable output, as discrete counts are automatically grounded in the image. Furthermore, our method outperforms the state of the art architecture for VQA on multiple metrics that evaluate counting. | INTERPRETABLE COUNTING FOR VISUAL QUESTION ANSWERING |
d259298593 | The generalization ability of machine learning models degrades significantly when the test distribution shifts away from the training distribution.We investigate the problem of training models that are robust to shifts caused by changes in the distribution of class-priors or group-priors.The presence of skewed training priors can often lead to the models overfitting to spurious features.Unlike existing methods, which optimize for either the worst or the average performance over classes or groups, our work is motivated by the need for finer control over the robustness properties of the model.We present an extremely lightweight post-hoc approach that performs scaling adjustments to predictions from a pre-trained model, with the goal of minimizing a distributionally robust loss around a chosen target distribution.These adjustments are computed by solving a constrained optimization problem on a validation set and applied to the model during test time.Our constrained optimization objective is inspired from a natural notion of robustness to controlled distribution shifts.Our method comes with provable guarantees and empirically makes a strong case for distributional robust post-hoc classifiers.An empirical implementation is available at https://github.com/weijiaheng/Drops. | DISTRIBUTIONALLY ROBUST POST-HOC CLASSIFIERS UNDER PRIOR SHIFTS |
d249395519 | How the brain performs credit assignment is a fundamental unsolved problem in neuroscience. Many 'biologically plausible' algorithms have been proposed, which compute gradients that approximate those computed by backpropagation (BP), and which operate in ways that more closely satisfy the constraints imposed by neural circuitry. Many such algorithms utilize the framework of energybased models (EBMs), in which all free variables in the model are optimized to minimize a global energy function. However, in the literature, these algorithms exist in isolation and no unified theory exists linking them together. Here, we provide a comprehensive theory of the conditions under which EBMs can approximate BP, which lets us unify many of the BP approximation results in the literature (namely, predictive coding, equilibrium propagation, and contrastive Hebbian learning) and demonstrate that their approximation to BP arises from a simple and general mathematical property * Corresponding author. arXiv:2206.02629v3 [cs.LG] 3 Aug 2022 of EBMs at free-phase equilibrium. This property can then be exploited in different ways with different energy functions, and these specific choices yield a family of BP-approximating algorithms, which both includes the known results in the literature and can be used to derive new ones. | BACKPROPAGATION AT THE INFINITESIMAL INFERENCE LIMIT OF ENERGY-BASED MODELS: UNIFYING PREDICTIVE CODING, EQUILIBRIUM PROPAGATION, AND CONTRASTIVE HEBBIAN LEARNING |
d3474156 | Reading comprehension is a challenging task, especially when executed across longer or across multiple evidence documents, where the answer is likely to reoccur. Existing neural architectures typically do not scale to the entire evidence, and hence, resort to selecting a single passage in the document (either via truncation or other means), and carefully searching for the answer within that passage. However, in some cases, this strategy can be suboptimal, since by focusing on a specific passage, it becomes difficult to leverage multiple mentions of the same answer throughout the document. In this work, we take a different approach by constructing lightweight models that are combined in a cascade to find the answer. Each submodel consists only of feed-forward networks equipped with an attention mechanism, making it trivially parallelizable. We show that our approach can scale to approximately an order of magnitude larger evidence documents and can aggregate information from multiple mentions of each answer candidate across the document. Empirically, our approach achieves state-of-the-art performance on both the Wikipedia and web domains of the TriviaQA dataset, outperforming more complex, recurrent architectures. | MULTI-MENTION LEARNING FOR READING COMPREHENSION WITH NEURAL CASCADES |
d220525844 | 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.Preprint. Under review. | Lifelong Learning of Compositional Structures |
d260505344 | Vision transformers (ViTs) process input images as sequences of patches via selfattention; a radically different architecture than convolutional neural networks (CNNs). This makes it interesting to study the adversarial feature space of ViT models and their transferability. In particular, we observe that adversarial patterns found via conventional adversarial attacks show very low black-box transferability even for large ViT models. We show that this phenomenon is only due to the sub-optimal attack procedures that do not leverage the true representation potential of ViTs. A deep ViT is composed of multiple blocks, with a consistent architecture comprising of self-attention and feed-forward layers, where each block is capable of independently producing a class token. Formulating an attack using only the last class token (conventional approach) does not directly leverage the discriminative information stored in the earlier tokens, leading to poor adversarial transferability of ViTs. Using the compositional nature of ViT models, we enhance transferability of existing attacks by introducing two novel strategies specific to the architecture of ViT models. (i) Self-Ensemble: We propose a method to find multiple discriminative pathways by dissecting a single ViT model into an ensemble of networks. This allows explicitly utilizing class-specific information at each ViT block. (ii) Token Refinement: We then propose to refine the tokens to further enhance the discriminative capacity at each block of ViT. Our token refinement systematically combines the class tokens with structural information preserved within the patch tokens. An adversarial attack when applied to such refined tokens within the ensemble of classifiers found in a single vision transformer has significantly higher transferability and thereby brings out the true generalization potential of the ViT's adversarial space. Code: https://t.ly/hBbW. | ON IMPROVING ADVERSARIAL TRANSFERABILITY OF VISION TRANSFORMERS |
d226964491 | Pre-trained language models (PTLM) have achieved impressive results in a range of natural language understanding (NLU) and generation (NLG) tasks. However, current pre-training objectives such as masked token prediction (for BERT-style PTLMs) and masked span infilling (for T5-style PTLMs) do not explicitly model the relational commonsense knowledge about everyday concepts, which is crucial to many downstream tasks that need common sense to understand or generate. To augment PTLMs with concept-centric commonsense knowledge, in this paper, we propose both generative and contrastive objectives for learning common sense from the text, and use them as intermediate self-supervised learning tasks for incrementally pre-training PTLMs (before task-specific fine-tuning on downstream datasets). Furthermore, we develop a joint pre-training framework to unify generative and contrastive objectives so that they can mutually reinforce each other. Extensive experimental results show that our method, concept-aware language model (CALM) 1 , can pack more commonsense knowledge into the parameters of a pre-trained text-to-text transformer without relying on external knowledge graphs, yielding better performance on both NLU and NLG tasks. We show that while only incrementally pre-trained on a relatively small corpus for a few steps, CALM outperforms baseline methods by a consistent margin and even comparable with some larger PTLMs, which suggests that CALM can serve as a general, "plug-and-play" method for improving the commonsense reasoning ability of a PTLM. * Equal contribution. The work was done when Wangchunshu was visiting USC. 1 Our code and data will be made public at: https://github.com/INK-USC/CALM arXiv:2011.07956v1 [cs.CL] 24 Oct 2020 Preprint. Work in progress.knowledge, prior works mainly focus on training larger models(Brown et al., 2020), adding specific architectures to exploit external knowledge(Peters et al., 2019), or incorporating knowledge bases for pre-training . In this paper, we instead look to explicitly teach pre-trained models to write and reason with common concepts through novel pre-training strategies. We present two kinds of self-supervised pre-training tasks: concept-to-sentence generation (C2S) and concept order recovering (COR). C2S trains the pre-trained model to compose ("write") sentences given a set of concepts, and expects the generated sentences to be fluent and plausible in terms of commonsense. COR aims to teach models to detect and revise a corrupted sentence with incorrect ordering of concepts. As illustrated inFigure 1, both tasks require a pre-trained model to recall relevant commonsense facts about the concepts and to understand the underlying commonsense relations between them. Both of the proposed objectives can explicitly encourage the model to capture the relational concept-centric commonsense knowledge and perform compositional reasoning. | PRE-TRAINING TEXT-TO-TEXT TRANSFORMERS FOR CONCEPT-CENTRIC COMMON SENSE |
d208513914 | For sequence models with large vocabularies, a majority of network parameters lie in the input and output layers. In this work, we describe a new method, DeFINE, for learning deep token representations efficiently. Our architecture uses a hierarchical structure with novel skip-connections which allows for the use of low dimensional input and output layers, reducing total parameters and training time while delivering similar or better performance versus existing methods. DeFINE can be incorporated easily in new or existing sequence models. Compared to state-of-the-art methods including adaptive input representations, this technique results in a 6% to 20% drop in perplexity. On WikiText-103, DeFINE reduces the total parameters of Transformer-XL by half with minimal impact on performance. On the Penn Treebank, DeFINE improves AWD-LSTM by 4 points with a 17% reduction in parameters, achieving comparable performance to state-of-the-art methods with fewer parameters. For machine translation, DeFINE improves the efficiency of the Transformer model by about 1.4 times while delivering similar performance. | DEFINE: DEEP FACTORIZED INPUT TOKEN EMBEDDINGS FOR NEURAL SEQUENCE MODELING |
d67877096 | Numerous past works have tackled the problem of task-driven navigation. But, how to effectively explore a new environment to enable a variety of down-stream tasks has received much less attention. In this work, we study how agents can autonomously explore realistic and complex 3D environments without the context of task-rewards. We propose a learning-based approach and investigate different policy architectures, reward functions, and training paradigms. We find that use of policies with spatial memory that are bootstrapped with imitation learning and finally finetuned with coverage rewards derived purely from on-board sensors can be effective at exploring novel environments. We show that our learned exploration policies can explore better than classical approaches based on geometry alone and generic learning-based exploration techniques. Finally, we also show how such task-agnostic exploration can be used for down-stream tasks. Code and Videos are available at: https://sites.google.com/view/exploration-for-nav/. | LEARNING EXPLORATION POLICIES FOR NAVIGATION |
d57373913 | In many real-world learning scenarios, features are only acquirable at a cost constrained under a budget. In this paper, we propose a novel approach for costsensitive feature acquisition at the prediction-time. The suggested method acquires features incrementally based on a context-aware feature-value function. We formulate the problem in the reinforcement learning paradigm, and introduce a reward function based on the utility of each feature. Specifically, MC dropout sampling is used to measure expected variations of the model uncertainty which is used as a feature-value function. Furthermore, we suggest sharing representations between the class predictor and value function estimator networks. The suggested approach is completely online and is readily applicable to stream learning setups. The solution is evaluated on three different datasets including the well-known MNIST dataset as a benchmark as well as two cost-sensitive datasets: Yahoo Learning to Rank and a dataset in the medical domain for diabetes classification. According to the results, the proposed method is able to efficiently acquire features and make accurate predictions.A version of the source code and the health dataset preproccessing code for this paper is available at: https | OPPORTUNISTIC LEARNING: BUDGETED COST-SENSITIVE LEARNING FROM DATA STREAMS |
d211003742 | Conventional Generative Adversarial Networks (GANs) for text generation tend to have issues of reward sparsity and mode collapse that affect the quality and diversity of generated samples. To address the issues, we propose a novel self-adversarial learning (SAL) paradigm for improving GANs' performance in text generation. In contrast to standard GANs that use a binary classifier as its discriminator to predict whether a sample is real or generated, SAL employs a comparative discriminator which is a pairwise classifier for comparing the text quality between a pair of samples. During training, SAL rewards the generator when its currently generated sentence is found to be better than its previously generated samples. This self-improvement reward mechanism allows the model to receive credits more easily and avoid collapsing towards the limited number of real samples, which not only helps alleviate the reward sparsity issue but also reduces the risk of mode collapse. Experiments on text generation benchmark datasets show that our proposed approach substantially improves both the quality and the diversity, and yields more stable performance compared to the previous GANs for text generation. | SELF-ADVERSARIAL LEARNING WITH COMPARATIVE DISCRIMINATION FOR TEXT GENERATION |
d253098907 | Deformable Object Manipulation (DOM) is of significant importance to both daily and industrial applications. Recent successes in differentiable physics simulators allow learning algorithms to train a policy with analytic gradients through environment dynamics, which significantly facilitates the development of DOM algorithms. However, existing DOM benchmarks are either single-object-based or non-differentiable. This leaves the questions of 1) how a task-specific algorithm performs on other tasks and 2) how a differentiable-physics-based algorithm compares with the non-differentiable ones in general. In this work, we present DaXBench, a differentiable DOM benchmark with a wide object and task coverage. DaXBench includes 9 challenging high-fidelity simulated tasks, covering rope, cloth, and liquid manipulation with various difficulty levels. To better understand the performance of general algorithms on different DOM tasks, we conduct comprehensive experiments over representative DOM methods, ranging from planning to imitation learning and reinforcement learning. In addition, we provide careful empirical studies of existing decision-making algorithms based on differentiable physics, and discuss their limitations, as well as potential future directions. The code and video can be found here 1 . * These authors contributed equally. This work is partially completed at the SEA AI Lab. 1 The link of the project is https://github.com/AdaCompNUS/DaXBench | DAXBENCH: BENCHMARKING DEFORMABLE OBJECT MANIPULATION WITH DIFFERENTIABLE PHYSICS |
d231749906 | As large-scale graphs become increasingly more prevalent, it poses significant computational challenges to process, extract and analyze large graph data. Graph coarsening is one popular technique to reduce the size of a graph while maintaining essential properties. Despite rich graph coarsening literature, there is only limited exploration of data-driven methods in the field. In this work, we leverage the recent progress of deep learning on graphs for graph coarsening. We first propose a framework for measuring the quality of coarsening algorithm and show that depending on the goal, we need to carefully choose the Laplace operator on the coarse graph and associated projection/lift operators. Motivated by the observation that the current choice of edge weight for the coarse graph may be suboptimal, we parametrize the weight assignment map with graph neural networks and train it to improve the coarsening quality in an unsupervised way. Through extensive experiments on both synthetic and real networks, we demonstrate that our method significantly improves common graph coarsening methods under various metrics, reduction ratios, graph sizes, and graph types. It generalizes to graphs of larger size (25× of training graphs), is adaptive to different losses (differentiable and non-differentiable), and scales to much larger graphs than previous work.Published as a conference paper at ICLR 2021 graph. This means the weights of the coarse graph is determined by the coarsening algorithm (of the vertex set), leaving no room for adjustment.With the two observations above, we aim to develop a data-driven approach to better assigning weights for the coarse graph depending on specific goals at hand. We will leverage the recent progress of deep learning on graphs to develop a framework to learn to assign edge weights in an unsupervised manner from a collection of input (small) graphs. This learned weight-assignment map can then be applied to new graphs (of potentially much larger sizes). In particular, our contributions are threefold.• First, depending on the quantity of interest F (such as the quadratic form w.r.t. Laplace operator), one has to carefully choose projection/lift operator to relate quantities defined on graphs of different sizes. We formulate this as the invariance of F under lift map, and provide three cases of projection/lift map as well as the corresponding operators on the coarse graph. Interestingly, those operators all can be seen as the special cases of doubly-weighted Laplace operators on coarse graphs (Horak & Jost, 2013). • Second, we are the first to propose and develop a framework to learn the edge weights of the coarse graphs via graph neural networks (GNN) in an unsupervised manner. We show convincing results both theoretically and empirically that changing the weights is crucial to improve the quality of coarse graphs. • Third, through extensive experiments on both synthetic graphs and real networks, we demonstrate that our method GOREN significantly improves common graph coarsening methods under different evaluation metrics, reduction ratios, graph sizes, and graph types. It generalizes to graphs of larger size (than the training graphs), adapts to different losses (so as to preserve different properties of original graphs), and scales to much larger graphs than what previous work can handle. Even for losses that are not differentiable w.r.t the weights of the coarse graph, we show training networks with a differentiable auxiliary loss still improves the result. | GRAPH COARSENING WITH NEURAL NETWORKS |
d231846526 | Model Agnostic Meta-Learning (MAML) has emerged as a standard framework for meta-learning, where a meta-model is learned with the ability of fast adapting to new tasks. However, as a double-looped optimization problem, MAML needs to differentiate through the whole inner-loop optimization path for every outer-loop training step, which may lead to both computational inefficiency and sub-optimal solutions. In this paper, we generalize MAML to allow meta-learning to be defined in function spaces, and propose the first meta-learning paradigm in the Reproducing Kernel Hilbert Space (RKHS) induced by the meta-model's Neural Tangent Kernel (NTK). Within this paradigm, we introduce two meta-learning algorithms in the RKHS, which no longer need a sub-optimal iterative inner-loop adaptation as in the MAML framework. We achieve this goal by 1) replacing the adaptation with a fast-adaptive regularizer in the RKHS; and 2) solving the adaptation analytically based on the NTK theory. Extensive experimental studies demonstrate advantages of our paradigm in both efficiency and quality of solutions compared to related meta-learning algorithms. Another interesting feature of our proposed methods is that they are demonstrated to be more robust to adversarial attacks and out-ofdistribution adaptation than popular baselines, as demonstrated in our experiments.Published as a conference paper at ICLR 2021 function adaptation: one replaces the function adaptation step in the inner-loop with a new metaobjective with a fast-adaptive regularizer inspired by MAML; the other solves the adaptation problem analytically based on tools from NTK so that the meta-objective can be directly evaluated on samples in a closed-form. When restricting the function space to be RKHS, the solutions to the proposed two algorithms become conveniently solvable. In addition, we provide theoretical analysis on our proposed algorithms in the cases of using fully-connected neural networks and convolutional neural networks as the meta-model. Our analysis shows close connections between our methods and the existing ones. Particularly, we prove that one of our algorithms is closely related to MAML with some high-order terms ignored in the meta-objective function, thus endowing effective optimization. In summary, our main contributions are:• We re-analyze the meta-learning problem and introduce two new algorithms for metalearning in RKHS. Different from all existing meta-learning algorithms, our proposed methods can be solved efficiently without cumbersome chain-based adaptations. • We conduct theoretically analysis on the proposed algorithms, which suggests that our proposed algorithms are closely related to the existing MAML methods when fully-connected neural networks and convolutional neural networks are used as the meta-model. • We conduct extensive experiments to validate our algorithms. Experimental results indicate the effectiveness of our proposed methods, through standard few-shot learning, robustness to adversarial attacks and out-of-distribution adaptation. Learning to balance: Bayesian meta-learning for imbalanced and out-of-distribution tasks. Towards deep learning models resistant to adversarial attacks. ArXiv, abs/1706.06083, 2017.M.Arioli, B.Codenotti, and C.Fassino. The P adé method for computing the matrix exponential. Linear Algebra and its Applications, June 1996.Nikhil Mishra, Mostafa Rohaninejad, Xi Chen, and Pieter Abbeel. A simple neural attentive meta- | META-LEARNING WITH NEURAL TANGENT KERNELS |
d53115163 | We introduce an exploration bonus for deep reinforcement learning methods that is easy to implement and adds minimal overhead to the computation performed. The bonus is the error of a neural network predicting features of the observations given by a fixed randomly initialized neural network. We also introduce a method to flexibly combine intrinsic and extrinsic rewards. We find that the random network distillation (RND) bonus combined with this increased flexibility enables significant progress on several hard exploration Atari games. In particular we establish state of the art performance on Montezuma's Revenge, a game famously difficult for deep reinforcement learning methods. To the best of our knowledge, this is the first method that achieves better than average human performance on this game without using demonstrations or having access to the underlying state of the game, and occasionally completes the first level. | EXPLORATION BY RANDOM NETWORK DISTILLATION |
d3278107 | Deep neural networks have advanced the state of the art in named entity recognition. However, under typical training procedures, advantages over classical methods emerge only with large datasets. As a result, deep learning is employed only when large public datasets or a large budget for manually labeling data is available. In this work, we show that by combining deep learning with active learning, we can outperform classical methods even with a significantly smaller amount of training data. | Deep Active Learning for Named Entity Recognition |
d251929117 | High-definition (HD) map provides abundant and precise environmental information of the driving scene, serving as a fundamental and indispensable component for planning in autonomous driving system. We present MapTR, a structured end-to-end Transformer for efficient online vectorized HD map construction. We propose a unified permutation-equivalent modeling approach, i.e., modeling map element as a point set with a group of equivalent permutations, which accurately describes the shape of map element and stabilizes the learning process. We design a hierarchical query embedding scheme to flexibly encode structured map information and perform hierarchical bipartite matching for map element learning. MapTR achieves the best performance and efficiency with only camera input among existing vectorized map construction approaches on nuScenes dataset. In particular, MapTR-nano runs at real-time inference speed (25.1 FPS) on RTX 3090, 8× faster than the existing state-of-the-art camera-based method while achieving 5.0 higher mAP. Even compared with the existing stateof-the-art multi-modality method, MapTR-nano achieves 0.7 higher mAP , and MapTR-tiny achieves 13.5 higher mAP and 3× faster inference speed. Abundant qualitative results show that MapTR maintains stable and robust map construction quality in complex and various driving scenes. MapTR is of great application value in autonomous driving. Code and more demos are available at https://github.com/hustvl/MapTR. | MAPTR: STRUCTURED MODELING AND LEARNING FOR ONLINE VECTORIZED HD MAP CONSTRUCTION |
d263608710 | We investigate the unsupervised constituency parsing task, which organizes words and phrases of a sentence into a hierarchical structure without using linguistically annotated data.We observe that existing unsupervised parsers capture differing aspects of parsing structures, which can be leveraged to enhance unsupervised parsing performance.To this end, we propose a notion of "tree averaging," based on which we further propose a novel ensemble method for unsupervised parsing.To improve inference efficiency, we further distill the ensemble knowledge into a student model; such an ensemble-then-distill process is an effective approach to mitigate the over-smoothing problem existing in common multi-teacher distilling methods.Experiments show that our method surpasses all previous approaches, consistently demonstrating its effectiveness and robustness across various runs, with different ensemble components, and under domain-shift conditions. 1 | Ensemble Distillation for Unsupervised Constituency Parsing |
d248266593 | We study the theory of neural network (NN) from the lens of classical nonparametric regression problems with a focus on NN's ability to adaptively estimate functions with heterogeneous smoothness -a property of functions in Besov or Bounded Variation (BV) classes. Existing work on this problem requires tuning the NN architecture based on the function spaces and sample sizes. We consider a "Parallel NN" variant of deep ReLU networks and show that the standard weight decay is equivalent to promoting the p -sparsity (0 < p < 1) of the coefficient vector of an end-to-end learned function bases, i.e., a dictionary. Using this equivalence, we further establish that by tuning only the weight decay, such Parallel NN achieves an estimation error arbitrarily close to the minimax rates for both the Besov and BV classes. Notably, it gets exponentially closer to minimax optimal as the NN gets deeper. Our research sheds new lights on why depth matters and how NNs are more powerful than kernel methods.Are NNs locally adaptive, i.e., optimal in learning functions with heterogeneous smoothness?This is a timely question to ask, partly because the bulk of recent theory of NN leverages its asymptotic Reproducing Kernel Hilbert Space (RKHS) in the overparameterized regime [19, 5, 1]. RKHSbased approaches, e.g., kernel ridge regression with any fixed kernels are suboptimal in estimating functions with heterogeneous smoothness[9]. Therefore, existing deep learning theory based on RKHS does not satisfactorily explain the advantages of neural networks over kernel methods.We build upon the recent work of Suzuki[36]and Parhi and Nowak [27] who provided encouraging first answers to the question above about the local adaptivity of NNs. Specifically, Parhi and Nowak [27, Theorem 8] showed that a two-layer truncated power function activated neural network with a non-standard regularization is equivalent to the locally adaptive regression splines (LARS) [22]. This connection implies that such non-standard NNs achieve the minimax rate for the (higher order) bounded variation (BV) classes. We provide a detailed discussion about this work in Section B. Suzuki[36]showed that multilayer ReLU DNNs can achieve minimax rate for the Besov class, but requires the width, depth and an artificially imposed sparsity-level of the DNN weights to be carefully calibrated according to parameters of the Besov class, thus is quite different from how DNNs are typically trained in practice.In this paper, we aim at addressing the same locally adaptivity question for a more commonly used neural network with standard weight decayed training.Parallel neural networks. We restrict our attention on a special network architecture called parallel neural network [16,13]which learns an ensemble of subnetworks -each being a multilayer ReLU DNNs. Parallel NNs have been shown to be more well-behaved both theoretically [16,48,14,13]and empirically[47,41]. Moreover, the idea of parallel NNs was used in many successful NN architectures such as SqueezeNet, ResNext and Inception (see[13]and the references therein).Weight decay. Weight decay is a common method in deep learning to reduce overfitting. Empirically, the regularizer is not necessarily explicit. Many tricks in deep learning, including early stopping[45], quantization [18], and dropout[42]have similar effect as weight decay. In this paper, we make no assumption on the training method thus there is no (implicit) regularizers apart from weight decay.Summary of results. Our main contributions are:1. We prove that the (standard) weight decay in training an L-layer parallel ReLU-activated neural network is equivalent to a sparse p penalty term (where p = 2/L) on the linear coefficients of a learned representation.2. We show that neural networks can approximate B-spline basis functions of any order without the need of choosing the order parameter manually. In other words, neural networks can adapt to functions of different order of smoothness, and even functions with different smoothness in different regions in their domain.3. We show that the estimation error of weight decayed parallel ReLU neural network decreases polynomially with the number of samples up to a constant error for estimating functions with heterogeneous smoothness in the both BV and Besov classes, and the exponential term in the error rate is close to the minimax rate. Notably, the method requires tuning only the weight decay parameter.4. We find that deeper models achieve closer to the optimal error rate. This result helps explain why deep neural networks can achieve better performance than shallow ones empirically.The above results separate NNs with any linear methods such as kernel ridge regression. To the best of our knowledge, we are the first to demonstrate that standard techniques ("weight decay" and ReLU activation) suffice for DNNs in achieving the optimal rates for estimating BV and Besov functions. | Deep Learning meets Nonparametric Regression: Are Weight-Decayed DNNs Locally Adaptive? |
d249063147 | Direct speech-to-speech translation (S2ST) with discrete units leverages recent progress in speech representation learning. Specifically, a sequence of discrete representations derived in a self-supervised manner are predicted from the model and passed to a vocoder for speech reconstruction, while still facing the following challenges: 1) Acoustic multimodality: the discrete units derived from speech with same content could be indeterministic due to the acoustic property (e.g., rhythm, pitch, and energy), which causes deterioration of translation accuracy; 2) high latency: current S2ST systems utilize autoregressive models which predict each unit conditioned on the sequence previously generated, failing to take full advantage of parallelism. In this work, we propose TranSpeech, a speech-to-speech translation model with bilateral perturbation. To alleviate the acoustic multimodal problem, we propose bilateral perturbation (BiP), which consists of the style normalization and information enhancement stages, to learn only the linguistic information from speech samples and generate more deterministic representations. With reduced multimodality, we step forward and become the first to establish a non-autoregressive S2ST technique, which repeatedly masks and predicts unit choices and produces high-accuracy results in just a few cycles. Experimental results on three language pairs demonstrate that BiP yields an improvement of 2.9 BLEU on average compared with a baseline textless S2ST model. Moreover, our parallel decoding shows a significant reduction of inference latency, enabling speedup up to 21.4x than autoregressive technique. | TRANSPEECH: SPEECH-TO-SPEECH TRANSLATION WITH BILATERAL PERTURBATION |
d237605219 | Trust region methods rigorously enabled reinforcement learning (RL) agents to learn monotonically improving policies, leading to superior performance on a variety of tasks. Unfortunately, when it comes to multi-agent reinforcement learning (MARL), the property of monotonic improvement may not simply apply; this is because agents, even in cooperative games, could have conflicting directions of policy updates. As a result, achieving a guaranteed improvement on the joint policy where each agent acts individually remains an open challenge. In this paper, we extend the theory of trust region learning to cooperative MARL. Central to our findings are the multi-agent advantage decomposition lemma and the sequential policy update scheme. Based on these, we develop Heterogeneous-Agent Trust Region Policy Optimisation (HATPRO) and Heterogeneous-Agent Proximal Policy Optimisation (HAPPO) algorithms. Unlike many existing MARL algorithms, HATRPO/HAPPO do not need agents to share parameters, nor do they need any restrictive assumptions on decomposibility of the joint value function. Most importantly, we justify in theory the monotonic improvement property of HATRPO/HAPPO. We evaluate the proposed methods on a series of Multi-Agent MuJoCo and StarCraftII tasks. Results show that HATRPO and HAPPO significantly outperform strong baselines such as IPPO, MAPPO and MADDPG on all tested tasks, thereby establishing a new state of the art. * First two authors contribute equally. Code is available at https://github.com/cyanrain7/ Trust-Region-Policy-Optimisation-in-Multi-Agent-Reinforcement-Learning. | TRUST REGION POLICY OPTIMISATION IN MULTI-AGENT REINFORCEMENT LEARNING |
d259262449 | In this paper, we explore the application of mean field theory, a technique from statistical physics, to deep metric learning and address the high training complexity commonly associated with conventional metric learning loss functions. By adapting mean field theory for deep metric learning, we develop an approach to design classification-based loss functions from pair-based ones, which can be considered complementary to the proxy-based approach. Applying the mean field theory to two pair-based loss functions, we derive two new loss functions, MeanFieldContrastive and MeanFieldClassWiseMultiSimilarity losses, with reduced training complexity. We extensively evaluate these derived loss functions on three image-retrieval datasets and demonstrate that our loss functions outperform baseline methods in two out of the three datasets.Preprint. Under review. | Mean Field Theory in Deep Metric Learning |
d238744509 | We propose a novel prediction interval (PI) method for uncertainty quantification, which addresses three major issues with the state-of-the-art PI methods. First, existing PI methods require retraining of neural networks (NNs) for every given confidence level and suffer from the crossing issue in calculating multiple PIs. Second, they usually rely on customized loss functions with extra sensitive hyperparameters for which fine tuning is required to achieve a well-calibrated PI. Third, they usually underestimate uncertainties of out-of-distribution (OOD) samples leading to over-confident PIs. Our PI3NN method calculates PIs from linear combinations of three NNs, each of which is independently trained using the standard mean squared error loss. The coefficients of the linear combinations are computed using root-finding algorithms to ensure tight PIs for a given confidence level. We theoretically prove that PI3NN can calculate PIs for a series of confidence levels without retraining NNs and it completely avoids the crossing issue. Additionally, PI3NN does not introduce any unusual hyperparameters resulting in a stable performance. Furthermore, we address OOD identification challenge by introducing an initialization scheme which provides reasonably larger PIs of the OOD samples than those of the in-distribution samples. Benchmark and real-world experiments show that our method outperforms several state-of-the-art approaches with respect to predictive uncertainty quality, robustness, and OOD samples identification. | PI3NN: OUT-OF-DISTRIBUTION-AWARE PREDICTION INTERVALS FROM THREE NEURAL NETWORKS |
d49416020 | Stability is a key aspect of data analysis. In many applications, the natural notion of stability is geometric, as illustrated for example in computer vision. Scattering transforms construct deep convolutional representations which are certified stable to input deformations. This stability to deformations can be interpreted as stability with respect to changes in the metric structure of the domain. In this work, we show that scattering transforms can be generalized to non-Euclidean domains using diffusion wavelets, while preserving a notion of stability with respect to metric changes in the domain, measured with diffusion maps. The resulting representation is stable to metric perturbations of the domain while being able to capture "high-frequency" information, akin to the Euclidean Scattering. | DIFFUSION SCATTERING TRANSFORMS ON GRAPHS |
d235490479 | Self-and mutually-exciting point processes are popular models in machine learning and statistics for dependent discrete event data. To date, most existing models assume stationary kernels (including the classical Hawkes processes) and simple parametric models. Modern applications with complex event data require more general point process models that can incorporate contextual information of the events, called marks, besides the temporal and location information. Moreover, such applications often require non-stationary models to capture more complex spatio-temporal dependence. To tackle these challenges, a key question is to devise a versatile influence kernel in the point process model. In this paper, we introduce a novel and general neural network-based non-stationary influence kernel with high expressiveness for handling complex discrete events data while providing theoretical performance guarantees. We demonstrate the superior performance of our proposed method compared with the state-of-the-art on synthetic and real data. | NEURAL SPECTRAL MARKED POINT PROCESSES |
d237491843 | We present a novel adaptive optimization algorithm for large-scale machine learning problems. Equipped with a low-cost estimate of local curvature and Lipschitz smoothness, our method dynamically adapts the search direction and step-size. The search direction contains gradient information preconditioned by a well-scaled diagonal preconditioning matrix that captures the local curvature information. Our methodology does not require the tedious task of learning rate tuning, as the learning rate is updated automatically without adding an extra hyperparameter. We provide convergence guarantees on a comprehensive collection of optimization problems, including convex, strongly convex, and nonconvex problems, in both deterministic and stochastic regimes. We also conduct an extensive empirical evaluation on standard machine learning problems, justifying our algorithm's versatility and demonstrating its strong performance compared to other start-of-theart first-order and second-order methods. arXiv:2109.05198v1 [cs.LG] | Doubly Adaptive Scaled Algorithm for Machine Learning Using Second-Order Information |
d256615445 | In this work, we define a diffusion-based generative model capable of both music synthesis and source separation by learning the score of the joint probability density of sources sharing a context. Alongside the classic total inference tasks (i.e., generating a mixture, separating the sources), we also introduce and experiment on the partial generation task of source imputation, where we generate a subset of the sources given the others (e.g., play a piano track that goes well with the drums). Additionally, we introduce a novel inference method for the separation task based on Dirac likelihood functions. We train our model on Slakh2100, a standard dataset for musical source separation, provide qualitative results in the generation settings, and showcase competitive quantitative results in the source separation setting. Our method is the first example of a single model that can handle both generation and separation tasks, thus representing a step toward general audio models. * Equal contribution. Listing order is random. † Shared last authorship.Preprint. Under review. | Multi-Source Diffusion Models for Simultaneous Music Generation and Separation |
d226236734 | In this paper, we explore connections between interpretable machine learning and learning theory through the lens of local approximation explanations. First, we tackle the traditional problem of performance generalization and bound the testtime accuracy of a model using a notion of how locally explainable it is. Second, we explore the novel problem of explanation generalization which is an important concern for a growing class of finite sample-based local approximation explanations. Finally, we validate our theoretical results empirically and show that they reflect what can be seen in practice. * Denotes equal contribution 1 This terminology is not to be confused with example-based explanations where the explanation itself is in the form of data instances rather than a function. | A LEARNING THEORETIC PERSPECTIVE ON LOCAL EXPLAINABILITY |
d231933756 | Back-translation (Sennrich et al., 2016)is an effective strategy to improve the performance of Neural Machine Translation (NMT) by generating pseudo-parallel data. However, several recent works have found that better translation quality of the pseudo-parallel data does not necessarily lead to better final translation models, while lower-quality but more diverse data often yields stronger results(Edunov et al., 2018). In this paper, we propose a novel method to generate pseudo-parallel data from a pre-trained back-translation model. Our method is a meta-learning algorithm which adapts a pre-trained back-translation model so that the pseudoparallel data it generates would train a forward-translation model to do well on a validation set. In our evaluations in both the standard datasets WMT En-De'14 and WMT En-Fr'14, as well as a multilingual translation setting, our method leads to significant improvements over strong baselines. 1 How are you today? Wie geht es dir haute? Backward model Forward model (t-1) gradient samples Forward model (t) Ground Truth Parallel Data How are you? Wie geht es dir? Cross Entropy Loss update backward model | META BACK-TRANSLATION |
d250408279 | In this paper, we leverage low-level compiler intermediate representations (IR) to improve code translation. Traditional transpilers rely on syntactic information and handcrafted rules, which limits their applicability and produces unnaturallooking code. Applying neural machine translation (NMT) approaches to code has successfully broadened the set of programs on which one can get a naturallooking translation. However, they treat the code as sequences of text tokens, and still do not differentiate well enough between similar pieces of code which have different semantics in different languages. The consequence is low quality translation, reducing the practicality of NMT, and stressing the need for an approach significantly increasing its accuracy. Here we propose to augment code translation with IRs, specifically LLVM IR, with results on the C++, Java, Rust, and Go languages. Our method improves upon the state of the art for unsupervised code translation, increasing the number of correct translations by 11% on average, and up to 79% for the Java → Rust pair with greedy decoding. We extend previous test sets for code translation, by adding hundreds of Go and Rust functions. Additionally, we train models with high performance on the problem of IR decompilation, generating programming source code from IR, and study using IRs as pivot for translation. * Equal contribution 1 https://en.wikipedia.org/wiki/Source-to-source_compiler 1 arXiv:2207.03578v5 [cs.PL] | CODE TRANSLATION WITH COMPILER REPRESENTATIONS |
d263609114 | This paper rigorously shows how over-parameterization dramatically changes the convergence behaviors of gradient descent (GD) for the matrix sensing problem, where the goal is to recover an unknown low-rank ground-truth matrix from near-isotropic linear measurements. First, we consider the symmetric setting with the symmetric parameterization where M * ∈ R n×n is a positive semi-definite unknown matrix of rank r ≪ n, and one uses a symmetric parameterization XX ⊤ to learn M * . Here, X ∈ R n×k with k > r is the factor matrix. We give a novel Ω 1/T 2 lower bound of randomly initialized GD for the over-parameterized case (k > r) where T is the number of iterations. This is in stark contrast to the exact-parameterization scenario (k = r) where the convergence rate is exp (−Ω (T )). Next, we study asymmetric setting where M * ∈ R n1×n2 is the unknown matrix of rank r ≪ min{n 1 , n 2 }, and one uses an asymmetric parameterization F G ⊤ to learn M * where F ∈ R n1×k and G ∈ R n2×k . Building on prior work, we give a global exact convergence result of randomly initialized GD for the exactparameterization case (k = r) with an exp (−Ω (T )) rate. Furthermore, we give the first global exact convergence result for the over-parameterization case (k > r) with an exp −Ω α 2 T rate where α is the initialization scale. This linear convergence result in the over-parameterization case is especially significant because one can apply the asymmetric parameterization to the symmetric setting to speed up from Ω 1/T 2 to linear convergence. Therefore, we identify a surprising phenomenon: asymmetric parameterization can exponentially speed up convergence. Equally surprising is our analysis that highlights the importance of imbalance between F and G. This is in sharp contrast to prior works which emphasize balance. We further give an example showing the dependency on α in the convergence rate is unavoidable in the worst case. On the other hand, we propose a novel method that only modifies one step of GD and obtains a convergence rate independent of α, recovering the rate in the exact-parameterization case. We provide empirical studies to verify our theoretical findings.We first consider the symmetric matrix sensing setting, where M * ∈ R n×n is a positive semidefinite matrix of rank r ≪ n. A standard approach is to use a factored form XX ⊤ to learn M * where X ∈ R n×k . We call this symmetric parameterization because XX ⊤ is always symmetric and positive semi-definite. We will also introduce the asymmetric parameterization soon. We call the case when k = r the exact-parameterization because the rank of XX ⊤ matches that of M * . However, in practice, r is often unknown, so one may choose some large enough k > r to ensure the expressiveness of XX ⊤ , and we call this case over-parameterization.We consider using gradient descent to minimize the standard L 2 loss for training: L tr (X) = Srinadh Bhojanapalli, Anastasios Kyrillidis, and Sujay Sanghavi. Dropping convexity for faster semi-definite optimization. In Conference on Learning Theory, pp. 530-582. PMLR, 2016. Emmanuel Candes and Benjamin Recht. Exact matrix completion via convex optimization. Communications of the ACM, 55(6):111-119, 2012.Emmanuel J Candes and Yaniv Plan. Tight oracle inequalities for low-rank matrix recovery from a minimal number of noisy random measurements. . Low-rank matrix recovery with structural incoherence for robust face recognition. regularization in learning deep homogeneous models: Layers are automatically balanced. Advances in neural information processing systems, 31, 2018a. . Gradient descent provably optimizes over-parameterized neural networks. arXiv preprint arXiv:1810.02054, 2018b. misspecification and the convergence rate of em. 2020.Cong Fang, Jason Lee, Pengkun Yang, and Tong Zhang. Modeling from features: a mean-field framework for over-parameterized deep neural networks. | How Over-Parameterization Slows Down Gradient Descent in Matrix Sensing: The Curses of Symmetry and Initialization |
d227162420 | Multi-goal reaching is an important problem in reinforcement learning needed to achieve algorithmic generalization. Despite recent advances in this field, current algorithms suffer from three major challenges: high sample complexity, learning only a single way of reaching the goals, and difficulties in solving complex motion planning tasks. In order to address these limitations, we introduce the concept of cumulative accessibility functions, which measure the reachability of a goal from a given state within a specified horizon. We show that these functions obey a recurrence relation, which enables learning from offline interactions. We also prove that optimal cumulative accessibility functions are monotonic in the planning horizon. Additionally, our method can trade off speed and reliability in goal-reaching by suggesting multiple paths to a single goal depending on the provided horizon. We evaluate our approach on a set of multi-goal discrete and continuous control tasks. We show that our method outperforms state-of-the-art goal-reaching algorithms in success rate, sample complexity, and path optimality. Our code is available at https://github.com/layer6ai-labs/CAE, and additional visualizations can be found at https | C-LEARNING: HORIZON-AWARE CUMULATIVE ACCESSIBILITY ESTIMATION |
d263830324 | Scaling up representations for images or text has been extensively investigated in the past few years and has led to revolutions in learning vision and language.However, scalable representation for 3D objects and scenes is relatively unexplored.In this work, we present Uni3D, a 3D foundation model to explore the unified 3D representation at scale.Uni3D uses a 2D initialized ViT end-to-end pretrained to align the 3D point cloud features with the image-text aligned features.Via the simple architecture and pretext task, Uni3D can leverage abundant 2D pretrained models as initialization and image-text aligned models as the target, unlocking the great potential of 2D models and scaling-up strategies to the 3D world.We efficiently scale up Uni3D to one billion parameters, and set new records on a broad range of 3D tasks, such as zero-shot classification, few-shot classification, open-world understanding and part segmentation.We show that the strong Uni3D representation also enables applications such as 3D painting and retrieval in the wild.We believe that Uni3D provides a new direction for exploring both scaling up and efficiency of the representation in 3D domain. | UNI3D: EXPLORING UNIFIED 3D REPRESENTATION AT SCALE |
d248157178 | Over the past few years, the acceleration of computing resources and research in deep learning has led to significant practical successes in a range of tasks, including in particular in computer vision. Building on these advances, reinforcement learning has also seen a leap forward with the emergence of agents capable of making decisions directly from visual observations. Despite these successes, the over-parametrization of neural architectures leads to memorization of the data used during training and thus to a lack of generalization. Reinforcement learning agents based on visual inputs also suffer from this phenomenon by erroneously correlating rewards with unrelated visual features such as background elements. To alleviate this problem, we introduce a new regularization technique consisting of channel-consistent local permutations (CLOP) of the feature maps. The proposed permutations induce robustness to spatial correlations and help prevent overfitting behaviors in RL. We demonstrate, on the OpenAI Procgen Benchmark, that RL agents trained with the CLOP method exhibit robustness to visual changes and better generalization properties than agents trained using other state-of-the-art regularization techniques. We also demonstrate the effectiveness of CLOP as a general regularization technique in supervised learning. | LOCAL FEATURE SWAPPING FOR GENERALIZATION IN REINFORCEMENT LEARNING |
d220280364 | Communication compression has been extensively adopted to speed up large-scale distributed optimization. However, most existing decentralized algorithms with compression are unsatisfactory in terms of convergence rate and stability. In this paper, we delineate two key obstacles in the algorithm design -data heterogeneity and compression error. Our attempt to explicitly overcome these obstacles leads to a novel decentralized algorithm named LEAD. This algorithm is the first LinEAr convergent Decentralized algorithm with communication compression. Our theory describes the coupled dynamics of the inaccurate model propagation and optimization process. We also provide the first consensus error bound without assuming bounded gradients. Empirical experiments validate our theoretical analysis and show that the proposed algorithm achieves state-of-the-art computation and communication efficiency.arXiv:2007.00232v1 [cs.LG] 1 Jul 2020• We delineate two key obstacles in the algorithm design, namely data heterogeneity and compression error, for communication compression in decentralized optimization, and our attempt to explicitly overcome these obstacles leads to a novel algorithm, LEAD.1. This work considers the non-stochastic setting only, where the full-batch gradient is used. For fairness, when comparing with other algorithms, we assume the availability of the full gradient, i.e., gradient variance is zero. Note the stochastic gradient can be used in LEAD, and it converges to the neighborhood of optimal solutions. We leave the stochastic version with variance reduction as future works. | Linear Convergent Decentralized Optimization with Compression |
d53192805 | Despite their impressive performance, deep neural networks exhibit striking failures on out-of-distribution inputs. One core idea of adversarial example research is to reveal neural network errors under such distribution shifts. We decompose these errors into two complementary sources: sensitivity and invariance. We show deep networks are not only too sensitive to task-irrelevant changes of their input, as is well-known from -adversarial examples, but are also too invariant to a wide range of task-relevant changes, thus making vast regions in input space vulnerable to adversarial attacks. We show such excessive invariance occurs across various tasks and architecture types. On MNIST and ImageNet one can manipulate the class-specific content of almost any image without changing the hidden activations. We identify an insufficiency of the standard cross-entropy loss as a reason for these failures. Further, we extend this objective based on an informationtheoretic analysis so it encourages the model to consider all task-dependent features in its decision. This provides the first approach tailored explicitly to overcome excessive invariance and resulting vulnerabilities. | EXCESSIVE INVARIANCE CAUSES ADVERSARIAL VULNERABILITY |
d263622509 | The recent success of Transformer in natural language processing has sparked its use in various domains.In offline reinforcement learning (RL), Decision Transformer (DT) is emerging as a promising model based on Transformer.However, we discovered that the attention module of DT is not appropriate to capture the inherent local dependence pattern in trajectories of RL modeled as a Markov decision process.To overcome the limitations of DT, we propose a novel action sequence predictor, named Decision ConvFormer (DC), based on the architecture of MetaFormer, which is a general structure to process multiple entities in parallel and understand the interrelationship among the multiple entities.DC employs local convolution filtering as the token mixer and can effectively capture the inherent local associations of the RL dataset.In extensive experiments, DC achieved stateof-the-art performance across various standard RL benchmarks while requiring fewer resources.Furthermore, we show that DC better understands the underlying meaning in data and exhibits enhanced generalization capability. | DECISION CONVFORMER: LOCAL FILTERING IN METAFORMER IS SUFFICIENT FOR DECISION MAKING |
d252918439 | Modern quantum annealers can find high-quality solutions to combinatorial optimisation objectives given as quadratic unconstrained binary optimisation (QUBO) problems. Unfortunately, obtaining suitable QUBO forms in computer vision remains challenging and currently requires problem-specific analytical derivations. Moreover, such explicit formulations impose tangible constraints on solution encodings. In stark contrast to prior work, this paper proposes to learn QUBO forms from data through gradient backpropagation instead of deriving them. As a result, the solution encodings can be chosen flexibly and compactly. Furthermore, our methodology is general and virtually independent of the specifics of the target problem type. We demonstrate the advantages of learnt QUBOs on the diverse problem types of graph matching, 2D point cloud alignment and 3D rotation estimation. Our results are competitive with the previous quantum state of the art while requiring much fewer logical and physical qubits, enabling our method to scale to larger problems. The code and the new dataset will be open-sourced. arXiv:2210.08114v1 [quant-ph] | QuAnt: Quantum Annealing with Learnt Couplings |
d246823998 | Conditional contrastive learning frameworks consider the conditional sampling procedure that constructs positive or negative data pairs conditioned on specific variables. Fair contrastive learning constructs negative pairs, for example, from the same gender (conditioning on sensitive information), which in turn reduces undesirable information from the learned representations; weakly supervised contrastive learning constructs positive pairs with similar annotative attributes (conditioning on auxiliary information), which in turn are incorporated into the representations. | CONDITIONAL CONTRASTIVE LEARNING WITH KERNEL |
d251953252 | We present a bi-encoder framework for named entity recognition (NER), which applies contrastive learning to map candidate text spans and entity types into the same vector representation space. Prior work predominantly approaches NER as sequence labeling or span classification. We instead frame NER as a representation learning problem that maximizes the similarity between the vector representations of an entity mention and its type. This makes it easy to handle nested and flat NER alike, and can better leverage noisy self-supervision signals. A major challenge to this bi-encoder formulation for NER lies in separating non-entity spans from entity mentions. Instead of explicitly labeling all non-entity spans as the same class Outside (O) as in most prior methods, we introduce a novel dynamic thresholding loss, learned in conjunction with the standard contrastive loss. Experiments show that our method performs well in both supervised and distantly supervised settings, for nested and flat NER alike, establishing new state of the art across standard datasets in the general domain (e.g., ACE2004, ACE2005, CoNLL2003) and high-value verticals such as biomedicine (e.g., GENIA, NCBI, BC5CDR, JNLPBA). We release the code at github.com/microsoft/binder. | OPTIMIZING BI-ENCODER FOR NAMED ENTITY RECOGNITION VIA CONTRASTIVE LEARNING |
d238408406 | Learning accurate classifiers for novel categories from very few examples, known as few-shot image classification, is a challenging task in statistical machine learning and computer vision. The performance in few-shot classification suffers from the bias in the estimation of classifier parameters; however, an effective underlying bias reduction technique that could alleviate this issue in training few-shot classifiers has been overlooked. In this work, we demonstrate the effectiveness of Firth bias reduction in few-shot classification. Theoretically, Firth bias reduction removes the O(N −1 ) first order term from the small-sample bias of the Maximum Likelihood Estimator. Here we show that the general Firth bias reduction technique simplifies to encouraging uniform class assignment probabilities for multinomial logistic classification, and almost has the same effect in cosine classifiers. We derive an easy-to-implement optimization objective for Firth penalized multinomial logistic and cosine classifiers, which is equivalent to penalizing the cross-entropy loss with a KL-divergence between the uniform label distribution and the predictions. Then, we empirically evaluate that it is consistently effective across the board for few-shot image classification, regardless of (1) the feature representations from different backbones, (2) the number of samples per class, and (3) the number of classes. Finally, we show the robustness of Firth bias reduction, in the case of imbalanced data distribution. Our implementation is available at | ON THE IMPORTANCE OF FIRTH BIAS REDUCTION IN FEW-SHOT CLASSIFICATION |
d199000713 | Many recent methods for unsupervised or self-supervised representation learning train feature extractors by maximizing an estimate of the mutual information (MI) between different views of the data. This comes with several immediate problems: For example, MI is notoriously hard to estimate, and using it as an objective for representation learning may lead to highly entangled representations due to its invariance under arbitrary invertible transformations. Nevertheless, these methods have been repeatedly shown to excel in practice. In this paper we argue, and provide empirical evidence, that the success of these methods might be only loosely attributed to the properties of MI, and that they strongly depend on the inductive bias in both the choice of feature extractor architectures and the parametrization of the employed MI estimators. Finally, we establish a connection to deep metric learning and argue that this interpretation may be a plausible explanation for the success of the recently introduced methods. * Equal contribution. Correspondence to Michael Tschannen (tschannen@google.com), Josip Djolonga (josipd@google.com), and Mario Lucic (lucic@google.com). | ON MUTUAL INFORMATION MAXIMIZATION FOR REPRESENTATION LEARNING |
d221802286 | Recent work has explored the possibility of pruning neural networks at initialization. We assess proposals for doing so: SNIP (Lee et al., 2019), GraSP (Wang et al., 2020), SynFlow (Tanaka et al., 2020, and magnitude pruning. Although these methods surpass the trivial baseline of random pruning, they remain below the accuracy of magnitude pruning after training, and we endeavor to understand why. We show that, unlike pruning after training, accuracy is the same or higher when randomly shuffling which weights these methods prune within each layer or sampling new initial values. As such, the per-weight pruning decisions made by these methods can be replaced by a per-layer choice of the fraction of weights to prune. This property undermines the claimed justifications for these methods and suggests broader challenges with the underlying pruning heuristics, the desire to prune at initialization, or both.Early pruning methods. We study the following methods for pruning early in training. We assess the fidelity of our reimplimentations of these methods in the appendices specified below.Random. This method issues each weight a random score s ∼ Uniform(0, 1) and removes weights with the lowest scores. Empirically, it prunes each layer to approximately sparsity s. Random pruning is a naive method for early pruning whose performance any new proposal should surpass. | PRUNING NEURAL NETWORKS AT INITIALIZATION: WHY ARE WE MISSING THE MARK? |
d198147763 | Decentralized training of deep learning models is a key element for enabling data privacy and on-device learning over networks, as well as for efficient scaling to large compute clusters. As current approaches suffer from limited bandwidth of the network, we propose the use of communication compression in the decentralized training context. We show that Choco-SGD-recently introduced and analyzed for strongly-convex objectives only-converges under arbitrary high compression ratio on general non-convex functions at the rate O 1/ √ nT where T denotes the number of iterations and n the number of workers. The algorithm achieves linear speedup in the number of workers and supports higher compression than previous state-of-the art methods. We demonstrate the practical performance of the algorithm in two key scenarios: the training of deep learning models (i) over distributed user devices, connected by a social network and (ii) in a datacenter (outperforming all-reduce time-wise). * Equal contribution.arXiv:1907.09356v1 [cs.LG] 22 Jul 2019As remedy, we translate the recently proposed Choco-SGD (Koloskova et al., 2019) algorithm to the non-convex setting and provide a concise convergence analysis. The method allows the use of a much richer class of compression operators (such as arbitrary low quantization, sparsification, etc.), but has only been analyzed in the strongly-convex setting so far.Contributions. Our contributions can be summarized as:• We show that Choco-SGD converges at rate O 1 / √ nT + n /(ρ 4 δ 2 T ) on non-convex smooth functions, where n denotes the number of nodes, T the number of iterations, ρ the spectral gap of the mixing matrix and δ the compression ratio. The main term, O 1 / √ nT , matches with the centralized baselines with exact communication and shows a linear speedup in the number of workers n. Both ρ and δ only affect the asymptotically smaller second term. Choco-SGD is the first decentralized algorithm that converges for arbitrary high compression (i.e. arbitrary small δ > 0) and achieves linear speedup.• On the practical side, we demonstrate the strong performance of the algorithm in two relevant scenarios, firstly for on-device training over a realistic distributed social network, where lowering the bandwidth requirements of joint training is especially impactful, and secondly for computational scalability of training deep learning models in a datacenter setting for resource efficiency and time-to-accuracy.• We are able to incorporate momentum into the decentralized algorithm, and analyze its practical performance. As generalization error is of utmost importance in deep learning applications, we demonstrate training as well as test error for our methods in comparison to centralized baselines. | Decentralized Deep Learning with Arbitrary Communication Compression |
d57189269 | Unsupervised image-to-image translation has gained considerable attention due to the recent impressive progress based on generative adversarial networks (GANs). However, previous methods often fail in challenging cases, in particular, when an image has multiple target instances and a translation task involves significant changes in shape, e.g., translating pants to skirts in fashion images. To tackle the issues, we propose a novel method, coined instance-aware GAN (InstaGAN), that incorporates the instance information (e.g., object segmentation masks) and improves multi-instance transfiguration. The proposed method translates both an image and the corresponding set of instance attributes while maintaining the permutation invariance property of the instances. To this end, we introduce a context preserving loss that encourages the network to learn the identity function outside of target instances. We also propose a sequential mini-batch inference/training technique that handles multiple instances with a limited GPU memory and enhances the network to generalize better for multiple instances. Our comparative evaluation demonstrates the effectiveness of the proposed method on different image datasets, in particular, in the aforementioned challenging cases. Code and results are available in https://github.com/sangwoomo/instagan. | INSTAGAN: INSTANCE-AWARE IMAGE-TO-IMAGE TRANSLATION |
d259088588 | Lip-to-speech involves generating a natural-sounding speech synchronized with a soundless video of a person talking. Despite recent advances, current methods still cannot produce high-quality speech with high levels of intelligibility for challenging and realistic datasets such as LRS3. In this work, we present LipVoicer, a novel method that generates high-quality speech, even for in-the-wild and rich datasets, by incorporating the text modality. Given a silent video, we first predict the spoken text using a pre-trained lip-reading network. We then condition a diffusion model on the video and use the extracted text through a classifier-guidance mechanism where a pre-trained automatic speech recognition (ASR) serves as the classifier. LipVoicer outperforms multiple lip-to-speech baselines on LRS2 and LRS3, which are in-the-wild datasets with hundreds of unique speakers in their test set and an unrestricted vocabulary. Moreover, our experiments show that the inclusion of the text modality plays a major role in the intelligibility of the produced speech, readily perceptible while listening, and is empirically reflected in the substantial reduction of the word error rate (WER) metric. We demonstrate the effectiveness of LipVoicer through human evaluation, which shows that it produces more natural and synchronized speech signals compared to competing methods. Finally, we created a demo showcasing LipVoicer's superiority in producing natural, synchronized, and intelligible speech, providing additional evidence of its effectiveness. Project page: https://lipvoicer.github.ioPreprint. Under review. | LipVoicer: Generating Speech from Silent Videos Guided by Lip Reading |
d259298460 | We consider monotone variational inequality (VI) problems in multi-GPU settings where multiple processors/workers/clients have access to local stochastic dual vectors. This setting includes a broad range of important problems from distributed convex minimization to min-max and games. Extra-gradient, which is a de facto algorithm for monotone VI problems, has not been designed to be communicationefficient. To this end, we propose a quantized generalized extra-gradient (Q-GenX), which is an unbiased and adaptive compression method tailored to solve VIs. We provide an adaptive step-size rule, which adapts to the respective noise profiles at hand and achieve a fast rate of O(1/T ) under relative noise, and an orderoptimal O(1/ √ T ) under absolute noise and show distributed training accelerates convergence. Finally, we validate our theoretical results by providing real-world experiments and training generative adversarial networks on multiple GPUs. * These authors contributed equally to this work.Published as a conference paper at ICLR 2023Published as a conference paper at ICLR 2023Algorithm 1 Q-GenX: Loops are executed in parallel on processors. At certain steps, each processor computes sufficient statistics of a parametric distribution to estimate distribution of dual vectors. Input: Local data, parameter vector (local copy) X t , Y t , learning rate {γ t }, and set of update steps U 1 for t = 1 to T do 2 if t ∈ U then 3 for i = 1 to K do 4 Compute sufficient statistics and update quantization levels ℓ tReceive c i,t from each processor i and decodeV i,settings. Intuitively, this improvement is explained by the fact that the noisy error measurements vanish while we approach a solution of VI. In Appendix J, we highlight random coordinate descent and random player updating as popular examples, which motivate Assumption 3. | DISTRIBUTED EXTRA-GRADIENT WITH OPTIMAL COMPLEXITY AND COMMUNICATION GUARANTEES |
d218486818 | We show how neural models can be used to realize piece-wise constant functions such as decision trees. The proposed architecture, which we call locally constant networks, builds on ReLU networks that are piece-wise linear and hence their associated gradients with respect to the inputs are locally constant. We formally establish the equivalence between the classes of locally constant networks and decision trees. Moreover, we highlight several advantageous properties of locally constant networks, including how they realize decision trees with parameter sharing across branching / leaves. Indeed, only M neurons suffice to implicitly model an oblique decision tree with 2 M leaf nodes. The neural representation also enables us to adopt many tools developed for deep networks (e.g., DropConnect (Wan et al., 2013)) while implicitly training decision trees. We demonstrate that our method outperforms alternative techniques for training oblique decision trees in the context of molecular property classification and regression tasks. 1 arXiv:1909.13488v2 [cs.LG] 3 May 2020 and Brooks, Monterey, CA, 1984. Leo Breiman. Random forests. Machine Learning, 45(1):5-32, 2001. Richard P Brent. Fast training algorithms for multilayer neural nets. IEEE Transactions on Neural Networks, 2(3):346-354, 1991. Miguel A Carreira-Perpinán and Pooya Tavallali. Alternating optimization of decision trees, with application to learning sparse oblique trees. In Advances in Neural Information Processing Systems, pp. 1211-1221, 2018. Krzysztof J Cios and Ning Liu. A machine learning method for generation of a neural network architecture: A continuous id3 algorithm. IEEE Transactions on Neural Networks, 3(2):280-291, 1992.Yilun Du and Igor Mordatch. Implicit generation and modeling with energy based models. In . Oc1: A randomized algorithm for building oblique decision trees. | OBLIQUE DECISION TREES FROM DERIVATIVES OF RELU NETWORKS |
d252781163 | Probabilistic Circuits (PCs) are a unified framework for tractable probabilistic models that support efficient computation of various probabilistic queries (e.g., marginal probabilities). One key challenge is to scale PCs to model large and highdimensional real-world datasets: we observe that as the number of parameters in PCs increases, their performance immediately plateaus. This phenomenon suggests that the existing optimizers fail to exploit the full expressive power of large PCs. We propose to overcome such bottleneck by latent variable distillation: we leverage the less tractable but more expressive deep generative models to provide extra supervision over the latent variables of PCs. Specifically, we extract information from Transformer-based generative models to assign values to latent variables of PCs, providing guidance to PC optimizers. Experiments on both image and language modeling benchmarks (e.g., ImageNet and WikiText-2) show that latent variable distillation substantially boosts the performance of large PCs compared to their counterparts without latent variable distillation. In particular, on the image modeling benchmarks, PCs achieve competitive performance against some of the widely-used deep generative models, including variational autoencoders and flow-based models, opening up new avenues for tractable generative modeling. * Authors contributed equally. | SCALING UP PROBABILISTIC CIRCUITS BY LATENT VARIABLE DISTILLATION |
d244714837 | Research on generalization bounds for deep networks seeks to give ways to predict test error using just the training dataset and the network parameters. While generalization bounds can give many insights about architecture design, training algorithms etc., what they do not currently do is yield good predictions for actual test error. A recently introduced Predicting Generalization in Deep Learning competition (Jiang et al., 2020) aims to encourage discovery of methods to better predict test error. The current paper investigates a simple idea: can test error be predicted using synthetic data, produced using a Generative Adversarial Network (GAN) that was trained on the same training dataset? Upon investigating several GAN models and architectures, we find that this turns out to be the case. In fact, using GANs pre-trained on standard datasets, the test error can be predicted without requiring any additional hyperparameter tuning. This result is surprising because GANs have well-known limitations (e.g. mode collapse) and are known to not learn the data distribution accurately. Yet the generated samples are good enough to substitute for test data. Several additional experiments are presented to explore reasons why GANs do well at this task. In addition to a new approach for predicting generalization, the counter-intuitive phenomena presented in our work may also call for a better understanding of GANs' strengths and limitations. | On Predicting Generalization using GANs |
d261531159 | We present a new accelerated stochastic second-order method that is robust to both gradient and Hessian inexactness, which occurs typically in machine learning. We establish theoretical lower bounds and prove that our algorithm achieves optimal convergence in both gradient and Hessian inexactness in this key setting. We further introduce a tensor generalization for stochastic higher-order derivatives. When the oracles are non-stochastic, the proposed tensor algorithm matches the global convergence of Nesterov Accelerated Tensor method. Both algorithms allow for approximate solutions of their auxiliary subproblems with verifiable conditions on the accuracy of the solution.Preprint. Under review. | Advancing the lower bounds: An accelerated, stochastic, second-order method with optimal adaptation to inexactness |
d227305788 | Learning task-agnostic dynamics models in high-dimensional observation spaces can be challenging for model-based RL agents. We propose a novel way to learn latent world models by learning to predict sequences of future actions conditioned on task completion. These task-conditioned models adaptively focus modeling capacity on task-relevant dynamics, while simultaneously serving as an effective heuristic for planning with sparse rewards. We evaluate our method on challenging visual goal completion tasks and show a substantial increase in performance compared to prior model-free approaches. arXiv:2012.02419v1 [cs.LG] 4 Dec 2020 Goal State L S T M ResNet ResNet Move Down Move Down Turn Left STOP Linear Linear Linear Linear Linear Linear Linear Linear | PLANNING FROM PIXELS USING INVERSE DYNAMICS MODELS |
d250407963 | In supervised learning, it has been shown that label noise in the data can be interpolated without penalties on test accuracy. We show that interpolating label noise induces adversarial vulnerability, and prove the first theorem showing the relationship between label noise and adversarial risk for any data distribution. Our results are almost tight if we do not make any assumptions on the inductive bias of the learning algorithm. We then investigate how different components of this problem affect this result including properties of the distribution. We also discuss non-uniform label noise distributions; and prove a new theorem showing uniform label noise induces nearly as large an adversarial risk as the worst poisoning with the same noise rate. Then, we provide theoretical and empirical evidence that uniform label noise is more harmful than typical real-world label noise. Finally, we show how inductive biases amplify the effect of label noise and argue the need for future work in this direction. * Equal contribution. | A LAW OF ADVERSARIAL RISK, INTERPOLATION, AND LABEL NOISE |
d88503668 | Despite the recent successes in robotic locomotion control, the design of robots, i.e., the design of their body structure, still heavily relies on human engineering. Automatic robot design has been a long studied subject, however, progress has been slow due to large combinatorial search space and the difficulty to efficiently evaluate the candidate structures. Note that one needs to both, search over many possible body structures, and choose among them based on how the robot with that structure performs in an environment. The latter means training an optimal controller given a candidate structure, which in itself is costly to obtain. In this paper, we propose Neural Graph Evolution (NGE), which performs evolutionary search in graph space, by iteratively evolving graph structures using simple mutation primitives. Key to our approach is to parameterize the control policies with graph neural networks, which allows us to transfer skills from previously evaluated designs during the graph search. This significantly reduces evaluation cost of new candidates and makes the search process orders of magnitude more efficient than that of past work. In addition, NGE applies Graph Mutation with Uncertainty (GM-UC) by incorporating model uncertainty, which reduces the search space by balancing exploration and exploitation. We show that NGE significantly outperforms previous methods in terms of convergence rate and final performance. As shown in experiments, NGE is the first algorithm that can automatically discover kinematically preferred robotic graph structures, such as a fish with two symmetric flat side-fins and a tail, or a cheetah with athletic front and back legs. NGE is extremely efficient, it finds plausible robotic structures within a day on a single 64 CPU-core Amazon EC2 machine. The code and project webpage are released 1 . * Two authors contribute equally. | NEURAL GRAPH EVOLUTION: TOWARDS EFFICIENT AUTOMATIC ROBOT DESIGN |
d222142276 | The dominant paradigm for learning video-text representations -noise contrastive learning -increases the similarity of the representations of pairs of samples that are known to be related, such as text and video from the same sample, and pushes away the representations of all other pairs. We posit that this last behaviour is too strict, enforcing dissimilar representations even for samples that are semantically-related -for example, visually similar videos or ones that share the same depicted action. In this paper, we propose a novel method that alleviates this by leveraging a generative model to naturally push these related samples together: each sample's caption must be reconstructed as a weighted combination of other support samples' visual representations. This simple idea ensures that representations are not overly-specialized to individual samples, are reusable across the dataset, and results in representations that explicitly encode semantics shared between samples, unlike noise contrastive learning. Our proposed method outperforms others by a large margin on MSR-VTT, VATEX and ActivityNet, for video-to-text and text-to-video retrieval. | Support-set bottlenecks for video-text representation learning |
d222133043 | We consider the problem of learning to communicate using multi-agent reinforcement learning (MARL). A common approach is to learn off-policy, using data sampled from a replay buffer. However, messages received in the past may not accurately reflect the current communication policy of each agent, and this complicates learning. We therefore introduce a 'communication correction' which accounts for the non-stationarity of observed communication induced by multi-agent learning. It works by relabelling the received message to make it likely under the communicator's current policy, and thus be a better reflection of the receiver's current environment. To account for cases in which agents are both senders and receivers, we introduce an ordered relabelling scheme. Our correction is computationally efficient and can be integrated with a range of off-policy algorithms. It substantially improves the ability of communicating MARL systems to learn across a variety of cooperative and competitive tasks.Since the introduction of deep Q-learning(Mnih et al., 2013), it has become very common to use previous online experience, for instance stored in a replay buffer, to train agents in an offline manner. An obvious difficulty with doing this is that the information concerned may be out of date, leading the agent woefully astray in cases where the environment of an agent changes over time. One obvious strategy is to discard old experiences. However, this is wasteful -it requires many more samples from the environment before adequate policies can be learned, and may prevent agents from leveraging past experience sufficiently to act in complex environments. Here, we consider an alternative, Orwellian possibility, of using present information to correct the past, showing that it can greatly improve an agent's ability to learn.We explore a paradigm case involving multiple agents that must learn to communicate to optimise their own or task-related objectives. As with deep Q-learning, modern model-free approaches often seek to learn this communication off-policy, using experience stored in a replay buffer (Foerster et al., 2016;Lowe et al., 2017;Peng et al., 2017). However, multi-agent reinforcement learning (MARL) can be particularly challenging as the underlying game-theoretic structure is well known to lead to non-stationarity, with past experience becoming obsolete as agents come progressively to use different communication codes. It is this that our correction addresses.Altering previously communicated messages is particularly convenient for our purposes as it has no direct effect on the actual state of the environment (Lowe et al., 2019), but a quantifiable effect on the observed message, which constitutes the receiver's 'social environment'. We can therefore determine what the received message would be under the communicator's current policy, rather than what it was when the experience was first generated. Once this is determined, we can simply relabel the past experience to better reflect the agent's current social environment, a form of off-environment correction (Ciosek & Whiteson, 2017).We apply our 'communication correction' using the framework of centralised training with decentralised control(Lowe et al., 2017;Foerster et al., 2018), in which extra information -in this case the policies and observations of other agents -is used during training to learn decentralised multi-1 arXiv:2010.01192v1 [cs.LG] 2 Oct 2020 agent policies. We show how it can be combined with existing off-policy algorithms, with little computational cost, to achieve strong performance in both the cooperative and competitive cases. | CORRECTING EXPERIENCE REPLAY FOR MULTI-AGENT COMMUNICATION |
d245216646 | Musical expression requires control of both what notes are played, and how they are performed. Conventional audio synthesizers provide detailed expressive controls, but at the cost of realism. Black-box neural audio synthesis and concatenative samplers can produce realistic audio, but have few mechanisms for control. In this work, we introduce MIDI-DDSP a hierarchical model of musical instruments that enables both realistic neural audio synthesis and detailed user control. Starting from interpretable Differentiable Digital Signal Processing (DDSP) synthesis parameters, we infer musical notes and high-level properties of their expressive performance (such as timbre, vibrato, dynamics, and articulation). This creates a 3-level hierarchy (notes, performance, synthesis) that affords individuals the option to intervene at each level, or utilize trained priors (performance given notes, synthesis given performance) for creative assistance. Through quantitative experiments and listening tests, we demonstrate that this hierarchy can reconstruct high-fidelity audio, accurately predict performance attributes for a note sequence, independently manipulate the attributes of a given performance, and as a complete system, generate realistic audio from a novel note sequence. By utilizing an interpretable hierarchy, with multiple levels of granularity, MIDI-DDSP opens the door to assistive tools to empower individuals across a diverse range of musical experience. 1 , et al. Language models are few-shot learners. arXiv preprint arXiv:Jukebox: A generative model for music. arXiv preprint arXiv:2005.00341, 2020.Sander Dieleman, Aaron van den Oord, and Karen Simonyan. The challenge of realistic music generation: modelling raw audio at scale. | MIDI-DDSP: DETAILED CONTROL OF MUSICAL PERFORMANCE VIA HIERARCHICAL MODELING |
d3508234 | We introduce a general method for improving the convergence rate of gradientbased optimizers that is easy to implement and works well in practice. We analyze the effectiveness of the method by applying it to stochastic gradient descent, stochastic gradient descent with Nesterov momentum, and Adam, showing that it improves upon these commonly used algorithms on a range of optimization problems; in particular the kinds of objective functions that arise frequently in deep neural network training. Our method works by dynamically updating the learning rate during optimization using the gradient with respect to the learning rate of the update rule itself. Computing this "hypergradient" needs little additional computation, requires only one extra copy of the original gradient to be stored in memory, and relies upon nothing more than what is provided by reverse-mode automatic differentiation. | Online Learning Rate Adaptation with Hypergradient Descent |
d208193111 | Adversarial training is one of the main defenses against adversarial attacks. In this paper, we provide the first rigorous study on diagnosing elements of large-scale adversarial training on ImageNet, which reveals two intriguing properties. First, we study the role of normalization. Batch Normalization (BN) is a crucial element for achieving state-of-the-art performance on many vision tasks, but we show it may prevent networks from obtaining strong robustness in adversarial training. One unexpected observation is that, for models trained with BN, simply removing clean images from training data largely boosts adversarial robustness, i.e., 18.3%. We relate this phenomenon to the hypothesis that clean images and adversarial images are drawn from two different domains. This two-domain hypothesis may explain the issue of BN when training with a mixture of clean and adversarial images, as estimating normalization statistics of this mixture distribution is challenging. Guided by this two-domain hypothesis, we show disentangling the mixture distribution for normalization, i.e., applying separate BNs to clean and adversarial images for statistics estimation, achieves much stronger robustness. Additionally, we find that enforcing BNs to behave consistently at training and testing can further enhance robustness. Second, we study the role of network capacity. We find our so-called "deep" networks are still shallow for the task of adversarial learning. Unlike traditional classification tasks where accuracy is only marginally improved by adding more layers to "deep" networks (e.g., ResNet-152), adversarial training exhibits a much stronger demand on deeper networks to achieve higher adversarial robustness. This robustness improvement can be observed substantially and consistently even by pushing the network capacity to an unprecedented scale, i.e., ResNet-638. | INTRIGUING PROPERTIES OF ADVERSARIAL TRAINING AT SCALE |
d251649165 | Graph Neural Networks (GNNs) are gaining extensive attention for their application in graph data. However, the black-box nature of GNNs prevents users from understanding and trusting the models, thus hampering their applicability. Whereas explaining GNNs remains a challenge, most existing methods fall into approximation based and perturbation based approaches with suffer from faithfulness problems and unnatural artifacts, respectively. To tackle these problems, we propose DEGREE (Decomposition based Explanation for GRaph nEural nEtworks) to provide a faithful explanation for GNN predictions. By decomposing the information generation and aggregation mechanism of GNNs, DEGREE allows tracking the contributions of specific components of the input graph to the final prediction. Based on this, we further design a subgraph level interpretation algorithm to reveal complex interactions between graph nodes that are overlooked by previous methods. The efficiency of our algorithm can be further improved by utilizing GNN characteristics. Finally, we conduct quantitative and qualitative experiments on synthetic and real-world datasets to demonstrate the effectiveness of DEGREE on node classification and graph classification tasks. | DEGREE: DECOMPOSITION BASED EXPLANATION FOR GRAPH NEURAL NETWORKS |
d1741724 | We study reinforcement learning of chatbots with recurrent neural network architectures when the rewards are noisy and expensive to obtain. For instance, a chatbot used in automated customer service support can be scored by quality assurance agents, but this process can be expensive, time consuming and noisy. Previous reinforcement learning work for natural language processing uses on-policy updates and/or is designed for on-line learning settings. We demonstrate empirically that such strategies are not appropriate for this setting and develop an off-policy batch policy gradient method (BPG). We demonstrate the efficacy of our method via a series of synthetic experiments and an Amazon Mechanical Turk experiment on a restaurant recommendations dataset. | BATCH POLICY GRADIENT METHODS FOR IMPROVING NEURAL CONVERSATION MODELS |
d238408001 | Vapnik-Chervonenkis (VC) theory has so far been unable to explain the small generalization error of overparametrized neural networks. Indeed, existing applications of VC theory to large networks obtain upper bounds on VC dimension that are proportional to the number of weights, and for a large class of networks, these upper bound are known to be tight. In this work, we focus on a class of partially quantized networks that we refer to as hyperplane arrangement neural networks (HANNs). Using a sample compression analysis, we show that HANNs can have VC dimension significantly smaller than the number of weights, while being highly expressive. In particular, empirical risk minimization over HANNs in the overparametrized regime achieves the minimax rate for classification with Lipschitz posterior class probability. We further demonstrate the expressivity of HANNs empirically. On a panel of 121 UCI datasets, overparametrized HANNs match the performance of state-of-the-art full-precision models. | VC DIMENSION OF PARTIALLY QUANTIZED NEURAL NETWORKS IN THE OVERPARAMETRIZED REGIME |
d263909212 | The reasoning capabilities of LLM (Large Language Model) are widely acknowledged in recent research, inspiring studies on tool learning and autonomous agents. LLM serves as the "brain" of agent, orchestrating multiple tools for collaborative multi-step task solving. Unlike methods invoking tools like calculators or weather APIs for straightforward tasks, multi-modal agents excel by integrating diverse AI models for complex challenges. However, current multi-modal agents neglect the significance of model selection: they primarily focus on the planning and execution phases, and will only invoke predefined task-specific models for each subtask, making the execution fragile. Meanwhile, other traditional model selection methods are either incompatible with or suboptimal for the multi-modal agent scenarios, due to ignorance of dependencies among subtasks arising by multi-step reasoning. To this end, we identify the key challenges therein and propose the M 3 framework as a plug-in with negligible runtime overhead at test-time. This framework improves model selection and bolsters the robustness of multi-modal agents in multi-step reasoning. In the absence of suitable benchmarks, we create MS-GQA, a new dataset specifically designed to investigate the model selection challenge in multi-modal agents. Our experiments reveal that our framework enables dynamic model selection, considering both user inputs and subtask dependencies, thereby robustifying the overall reasoning process. Our code and benchmark: https://github.com/LINs-lab/M3. | TOWARDS ROBUST MULTI-MODAL REASONING VIA MODEL SELECTION |
d202230734 | Fine-tuning a deep network trained with the standard cross-entropy loss is a strong baseline for few-shot learning. When fine-tuned transductively, this outperforms the current state-of-the-art on standard datasets such as Mini-Imagenet, Tiered-Imagenet, CIFAR-FS and FC-100 with the same hyper-parameters. The simplicity of this approach enables us to demonstrate the first few-shot learning results on the Imagenet-21k dataset. We find that using a large number of meta-training classes results in high few-shot accuracies even for a large number of few-shot classes. We do not advocate our approach as the solution for few-shot learning, but simply use the results to highlight limitations of current benchmarks and few-shot protocols. We perform extensive studies on benchmark datasets to propose a metric that quantifies the "hardness" of a few-shot episode. This metric can be used to report the performance of few-shot algorithms in a more systematic way. | A BASELINE FOR FEW-SHOT IMAGE CLASSIFICATION |
d245334863 | We formally map the problem of sampling from an unknown distribution with a density in R d to the problem of learning and sampling a smoother density in R M d obtained by convolution with a fixed factorial kernel: the new density is referred to as M-density and the kernel as multimeasurement noise model (MNM). The M-density in R M d is smoother than the original density in R d , easier to learn and sample from, yet for large M the two problems are mathematically equivalent since clean data can be estimated exactly given a multimeasurement noisy observation using the Bayes estimator. To formulate the problem, we derive the Bayes estimator for Poisson and Gaussian MNMs in closed form in terms of the unnormalized M-density. This leads to a simple least-squares objective for learning parametric energy and score functions. We present various parametrization schemes of interest including one in which studying Gaussian M-densities directly leads to multidenoising autoencoders-this is the first theoretical connection made between denoising autoencoders and empirical Bayes in the literature. Samples in R d are obtained by walk-jump sampling (Saremi & Hyvärinen, 2019) via underdamped Langevin MCMC (walk) to sample from M-density and the multimeasurement Bayes estimation (jump). We study permutation invariant Gaussian M-densities on MNIST, CIFAR-10, and FFHQ-256 datasets, and demonstrate the effectiveness of this framework for realizing fast-mixing stable Markov chains in high dimensions. arXiv:2112.09822v2 [stat.ML] 16 Jun 2022 Published as a conference paper at ICLR 2022 Software All models were implemented in the CPython (v3.8) library PyTorch v1.9.0 (Paszke et al., 2017) using the PyTorch Lightning framework v1.4.6 (Falcon et al.). | MULTIMEASUREMENT GENERATIVE MODELS |
d52891534 | The behavioral dynamics of multi-agent systems have a rich and orderly structure, which can be leveraged to understand these systems, and to improve how artificial agents learn to operate in them. Here we introduce Relational Forward Models (RFM) for multi-agent learning, networks that can learn to make accurate predictions of agents' future behavior in multi-agent environments. Because these models operate on the discrete entities and relations present in the environment, they produce interpretable intermediate representations which offer insights into what drives agents' behavior, and what events mediate the intensity and valence of social interactions. Furthermore, we show that embedding RFM modules inside agents results in faster learning systems compared to non-augmented baselines. As more and more of the autonomous systems we develop and interact with become multi-agent in nature, developing richer analysis tools for characterizing how and why agents make decisions is increasingly necessary. Moreover, developing artificial agents that quickly and safely learn to coordinate with one another, and with humans in shared environments, is crucial. | RELATIONAL FORWARD MODELS FOR MULTI-AGENT LEARNING |
d252222320 | Effective scaling and a flexible task interface enable large language models to excel at many tasks. We present PaLI (Pathways Language and Image model), a model that extends this approach to the joint modeling of language and vision. PaLI generates text based on visual and textual inputs, and with this interface performs many vision, language, and multimodal tasks, in many languages. To train PaLI, we make use of large pre-trained encoder-decoder language models and Vision Transformers (ViTs). This allows us to capitalize on their existing capabilities and leverage the substantial cost of training them. We find that joint scaling of the vision and language components is important. Since existing Transformers for language are much larger than their vision counterparts, we train a large, 4-billion parameter ViT (ViT-e) to quantify the benefits from even larger-capacity vision models. To train PaLI, we create a large multilingual mix of pre-training tasks, based on a new image-text training set containing 10B images and texts in over 100 languages. PaLI achieves state-of-the-art in multiple vision and language tasks (such as captioning, visual question-answering, scene-text understanding), while retaining a simple, modular, and scalable design. . MaXM: Towards multilingual visual question answering. arXiv preprint arXiv:2209.05401, 2022b. . Microsoft COCO Captions: Data collection and evaluation server. arXiv preprint arXiv:1504.00325, 2015. . Winner team Mia at TextVQA challenge 2021: Vision-and-language representation learning with pre-trained sequence-to-sequence model. arXiv preprint arXiv:2106.15332, 2021. , et al. Learning transferable visual models from natural language supervision. . Scaling vision with sparse mixture of experts. , et al. Scaling up models and data with t5x and seqio. arXiv preprint arXiv:2203.17189, 2022. Sun. Objects365: A large-scale, high-quality dataset for object detection. In . Learning robust global representations by penalizing local predictive power. | PALI: A JOINTLY-SCALED MULTILINGUAL LANGUAGE-IMAGE MODEL |
d247741535 | State-of-the-art deep Q-learning methods update Q-values using state transition tuples sampled from the experience replay buffer. This strategy often uniformly and randomly samples or prioritizes data sampling based on measures such as the temporal difference (TD) error. Such sampling strategies can be inefficient at learning Q-function because a state's Q-value depends on the Q-value of successor states. If the data sampling strategy ignores the precision of Q-value estimate of the next state, it can lead to useless and often incorrect updates to the Q-values. To mitigate this issue, we organize the agent's experience into a graph that explicitly tracks the dependency between Q-values of states. Each edge in the graph represents a transition between two states by executing a single action. We perform value backups via a breadth-first search starting from the set of terminal states and successively moving backwards. We empirically show that our method is substantially more data-efficient than several baselines on a diverse range of goal-reaching tasks. Notably, the proposed method also outperforms baselines that consume more batches of training experience and operates from high-dimensional observational data such as images. The code is available at: https://github.com/Improbable-AI/ter. | TOPOLOGICAL EXPERIENCE REPLAY |
d245335427 | We propose a framework to continuously learn object-centric representations for visual learning and understanding. Existing object-centric representations either rely on supervisions that individualize objects in the scene, or perform unsupervised disentanglement that can hardly deal with complex scenes in the real world. To mitigate the annotation burden and relax the constraints on the statistical complexity of the data, our method leverages interactions to effectively sample diverse variations of an object and the corresponding training signals while learning the object-centric representations. Throughout learning, objects are streamed one by one in random order with unknown identities, and are associated with latent codes that can synthesize discriminative weights for each object through a convolutional hypernetwork. Moreover, re-identification of learned objects and forgetting prevention are employed to make the learning process efficient and robust. We perform an extensive study of the key features of the proposed framework and analyze the characteristics of the learned representations. Furthermore, we demonstrate the capability of the proposed framework in learning representations that can improve label efficiency in downstream tasks. Our code and trained models are made publicly available at: https://github.com/pptrick/ Object-Pursuit. * Equal Contribution.arXiv:2112.07954v3 [cs.CV] 3 Apr 2022Published as a conference paper at ICLR 2022We employ network weights as the object-centric representations. Specifically, the proposed method learns an object-centric representation from the data collected by manipulating a single object, through learning a latent code that can be translated into a neural network. The neural network is produced by a discriminative weight generation hypernetwork and is able to distinguish the represented object from anything else. In order to learn representations for objects that stream in one by one, the proposed framework is augmented with an object re-identification procedure to avoid learning seen objects. Moreover, we hypothesize that object representations are embedded in a low-dimensional manifold, so the proposed framework first checks whether a new object can be represented by learned objects; if not, the new object will be learned as a base object serving the purpose of representing future objects, thus the name object pursuit. Furthermore, the proposed framework deals with the catastrophic forgetting of learned object representations by enforcing the hypernetwork to maintain the mapping between the learned representations and their corresponding network weights.In summary, our work makes the following contributions: 1) we propose a novel framework named object pursuit that can continuously learn object-centric representations using training data collected from interactions with individual objects, 2) we perform an extensive study to understand the pursuit dynamics and characterize its typical behaviors regarding the key design features, and 3) we analyze the learned object space, in terms of its succinctness and effectiveness in representing objects, and empirically demonstrate its potential for label efficient visual learning. learning. arXiv preprint arXiv:1710.10368, 2017.Tero Karras, Samuli Laine, and Timo Aila. A style-based generator architecture for generative adversarial networks. In . Fearnet: Brain-inspired model for incremental learning. | OBJECT PURSUIT: BUILDING A SPACE OF OBJECTS VIA DISCRIMINATIVE WEIGHT GENERATION |
d182952687 | Imitation learning allows agents to learn complex behaviors from demonstrations. However, learning a complex vision-based task may require an impractical number of demonstrations. Meta-imitation learning is a promising approach towards enabling agents to learn a new task from one or a few demonstrations by leveraging experience from learning similar tasks. In the presence of task ambiguity or unobserved dynamics, demonstrations alone may not provide enough information; an agent must also try the task to successfully infer a policy. In this work, we propose a method that can learn to learn from both demonstrations and trial-anderror experience with sparse reward feedback. In comparison to meta-imitation, this approach enables the agent to effectively and efficiently improve itself autonomously beyond the demonstration data. In comparison to meta-reinforcement learning, we can scale to substantially broader distributions of tasks, as the demonstration reduces the burden of exploration. Our experiments show that our method significantly outperforms prior approaches on a set of challenging, vision-based control tasks. | WATCH, TRY, LEARN: META-LEARNING FROM DEMONSTRATIONS AND REWARDS |
d210839350 | We present Decentralized Distributed Proximal Policy Optimization (DD-PPO), a method for distributed reinforcement learning in resource-intensive simulated environments.DD-PPO is distributed (uses multiple machines), decentralized (lacks a centralized server), and synchronous (no computation is ever 'stale'), making it conceptually simple and easy to implement.In our experiments on training virtual robots to navigate in Habitat-Sim(Savva et al., 2019), DD-PPO exhibits near-linear scaling -achieving a speedup of 107x on 128 GPUs over a serial implementation.We leverage this scaling to train an agent for 2.5 Billion steps of experience (the equivalent of 80 years of human experience) -over 6 months of GPU-time training in under 3 days of wall-clock time with 64 GPUs.This massive-scale training not only sets the state of art on Habitat Autonomous Navigation Challenge 2019, but essentially 'solves' the task -near-perfect autonomous navigation in an unseen environment without access to a map, directly from an RGB-D camera and a GPS+Compass sensor.Fortuitously, error vs computation exhibits a power-law-like distribution; thus, 90% of peak performance is obtained relatively early (at 100 million steps) and relatively cheaply (under 1 day with 8 GPUs).Finally, we show that the scene understanding and navigation policies learned can be transferred to other navigation tasks -the analog of 'ImageNet pre-training + task-specific fine-tuning' for embodied AI.Our model outperforms ImageNet pre-trained CNNs on these transfer tasks and can serve as a universal resource (all models and code are publicly available). | DD-PPO: LEARNING NEAR-PERFECT POINTGOAL NAVIGATORS FROM 2.5 BILLION FRAMES |
d3005102 | We introduce an exceptionally simple gated recurrent neural network (RNN) that achieves performance comparable to well-known gated architectures, such as LSTMs and GRUs, on the word-level language modeling task. We prove that our model has simple, predicable and non-chaotic dynamics. This stands in stark contrast to more standard gated architectures, whose underlying dynamical systems exhibit chaotic behavior. | A RECURRENT NEURAL NETWORK WITHOUT CHAOS |
d256615740 | Learning on graphs, where instance nodes are inter-connected, has become one of the central problems for deep learning, as relational structures are pervasive and induce data inter-dependence which hinders trivial adaptation of existing approaches that assume inputs to be i.i.d. sampled. However, current models mostly focus on improving testing performance of in-distribution data and largely ignore the potential risk w.r.t. out-of-distribution (OOD) testing samples that may cause negative outcome if the prediction is overconfident on them. In this paper, we investigate the under-explored problem, OOD detection on graph-structured data, and identify a provably effective OOD discriminator based on an energy function directly extracted from graph neural networks trained with standard classification loss. This paves a way for a simple, powerful and efficient OOD detection model for GNN-based learning on graphs, which we call GNNSAFE. It also has nice theoretical properties that guarantee an overall distinguishable margin between the detection scores for in-distribution and OOD samples, which, more critically, can be further strengthened by a learning-free energy belief propagation scheme. For comprehensive evaluation, we introduce new benchmark settings that evaluate the model for detecting OOD data from both synthetic and real distribution shifts (crossdomain graph shifts and temporal graph shifts). The results show that GNNSAFE achieves up to 17.0% AUROC improvement over state-of-the-arts and it could serve as simple yet strong baselines in such an under-developed area. The codes are available at https://github.com/qitianwu/GraphOOD-GNNSafe.Published as a conference paper at ICLR 2023 labels, in input graphs to learn stable predictive relations across environments, whileYang et al. (2022b)resorts to an analogy of thermodynamics diffusion on graphs to build a principled knowledge distillation model for geometric knowledge transfer and generalization.Out-of-Distribution Detection. One critical challenge that stands in the way for trustworthy AI systems is how to equip the deep learning models with enough reliability of realizing what they don't know, i.e., detecting OOD data on which the models are expected to have low confidence(Amodei et al., 2016;Liang et al., 2018). This is fairly important when it comes to safety-critical applications such as medical diagnosis(Kukar, 2003), autonomous driving(Dai & Van Gool, 2018), etc. Such a problem is called out-of-distribution detection in the literature, which aims at discriminating OOD data from the in-distribution one. While there are a surge of recent works exploring various effective methods for OOD detection(Hendrycks & Gimpel, 2016;Bevandić et al., 2018;DeVries & Taylor, 2018;Hein et al., 2019;Hsu et al., 2020;Sun et al., 2021;Bitterwolf et al., 2022), these models again tacitly assume inputs to be independently sampled and are hard to be applied for graph data.In this paper, we investigate into out-of-distribution detection for learning on graphs, where the model needs to particularly handle data inter-dependence induced by the graph. Our methodology is built on graph neural networks (GNNs) as an encoder for node representation/prediction that accommodates the structural information. As an important step to enhance the reliability of GNN models against OOD testing instances, we identify an intrinsic OOD discriminator from a GNN classifier trained with standard learning objective. The key insight of our work is that standard GNN classifiers possess inherent good capability for detecting OOD samples (i.e., what they don't know) from unknown testing observations, with details of the model as described below. | ENERGY-BASED OUT-OF-DISTRIBUTION DETECTION FOR GRAPH NEURAL NETWORKS |
d224705447 | We propose methods to strengthen the invariance properties of representations obtained by contrastive learning. While existing approaches implicitly induce a degree of invariance as representations are learned, we look to more directly enforce invariance in the encoding process. To this end, we first introduce a training objective for contrastive learning that uses a novel regularizer to control how the representation changes under transformation. We show that representations trained with this objective perform better on downstream tasks and are more robust to the introduction of nuisance transformations at test time. Second, we propose a change to how test time representations are generated by introducing a feature averaging approach that combines encodings from multiple transformations of the original input, finding that this leads to across the board performance gains. Finally, we introduce the novel Spirograph dataset to explore our ideas in the context of a differentiable generative process with multiple downstream tasks, showing that our techniques for learning invariance are highly beneficial. * Equal contribution 1 arXiv:2010.09515v1 [cs.LG] | IMPROVING TRANSFORMATION INVARIANCE IN CONTRASTIVE REPRESENTATION LEARNING |
d231632806 | We define a notion of information that an individual sample provides to the training of a neural network, and we specialize it to measure both how much a sample informs the final weights and how much it informs the function computed by the weights. Though related, we show that these quantities have a qualitatively different behavior. We give efficient approximations of these quantities using a linearized network and demonstrate empirically that the approximation is accurate for real-world architectures, such as pre-trained ResNets. We apply these measures to several problems, such as dataset summarization, analysis of under-sampled classes, comparison of informativeness of different data sources, and detection of adversarial and corrupted examples. Our work generalizes existing frameworks but enjoys better computational properties for heavily overparametrized models, which makes it possible to apply it to real-world networks. | ESTIMATING INFORMATIVENESS OF SAMPLES WITH SMOOTH UNIQUE INFORMATION |
d52954637 | The smallest eigenvectors of the graph Laplacian are well-known to provide a succinct representation of the geometry of a weighted graph. In reinforcement learning (RL), where the weighted graph may be interpreted as the state transition process induced by a behavior policy acting on the environment, approximating the eigenvectors of the Laplacian provides a promising approach to state representation learning. However, existing methods for performing this approximation are ill-suited in general RL settings for two main reasons: First, they are computationally expensive, often requiring operations on large matrices. Second, these methods lack adequate justification beyond simple, tabular, finite-state settings. In this paper, we present a fully general and scalable method for approximating the eigenvectors of the Laplacian in a model-free RL context. We systematically evaluate our approach and empirically show that it generalizes beyond the tabular, finite-state setting. Even in tabular, finite-state settings, its ability to approximate the eigenvectors outperforms previous proposals. Finally, we show the potential benefits of using a Laplacian representation learned using our method in goal-achieving RL tasks, providing evidence that our technique can be used to significantly improve the performance of an RL agent. | The Laplacian in RL: Learning Representations with Efficient Approximations |
d252600018 | Recent works have widely adopted large language model pretraining for source code, suggested source code-specific pretraining objectives and investigated the applicability of various Transformer-based language model architectures for source code. This work investigates another important aspect of such models, namely the effect of different subtokenization options, and aims at identifying most effective and length-efficient subtokenizations, taking into account code specifics. We propose subtokenziation that reduces average length by 17% without downstream performance drop, and show that a carefully chosen subtokenization may improve quality by 0.5-2%, possibly with some length increase. | CODEBPE: INVESTIGATING SUBTOKENIZATION OPTIONS FOR LARGE LANGUAGE MODEL PRETRAINING ON SOURCE CODE |
d52894096 | Recurrent neural networks (RNNs) have shown excellent performance in processing sequence data. However, they are both complex and memory intensive due to their recursive nature. These limitations make RNNs difficult to embed on mobile devices requiring real-time processes with limited hardware resources. To address the above issues, we introduce a method that can learn binary and ternary weights during the training phase to facilitate hardware implementations of RNNs. As a result, using this approach replaces all multiply-accumulate operations by simple accumulations, bringing significant benefits to custom hardware in terms of silicon area and power consumption. On the software side, we evaluate the performance (in terms of accuracy) of our method using long short-term memories (LSTMs) on various sequential models including sequence classification and language modeling. We demonstrate that our method achieves competitive results on the aforementioned tasks while using binary/ternary weights during the runtime. On the hardware side, we present custom hardware for accelerating the recurrent computations of LSTMs with binary/ternary weights. Ultimately, we show that LSTMs with binary/ternary weights can achieve up to 12× memory saving and 10× inference speedup compared to the full-precision implementation on an ASIC platform. | LEARNING RECURRENT BINARY/TERNARY WEIGHTS |
d233740346 | We study how representation learning can improve the efficiency of bandit problems. We study the setting where we play T linear bandits with dimension d concurrently, and these T bandit tasks share a common k(≪ d) dimensional linear representation. For the finite-action setting, we present a new algorithm whichwhere N is the number of rounds we play for each bandit. When T is sufficiently large, our algorithm significantly outperforms the naive algorithm (playing T bandits independently) that achieves O(T √ dN ) regret. We also provide an Ω(T √ kN + √ dkN T ) regret lower bound, showing that our algorithm is minimax-optimal up to poly-logarithmic factors. Furthermore, we extend our algorithm to the infinite-action setting and obtain a corresponding regret bound which demonstrates the benefit of representation learning in certain regimes. We also present experiments on synthetic and realworld data to illustrate our theoretical findings and demonstrate the effectiveness of our proposed algorithms. | IMPACT OF REPRESENTATION LEARNING IN LINEAR BANDITS |
d49303347 | Learning multimodal representations is a fundamentally complex research problem due to the presence of multiple heterogeneous sources of information. Although the presence of multiple modalities provides additional valuable information, there are two key challenges to address when learning from multimodal data: 1) models must learn the complex intra-modal and cross-modal interactions for prediction and 2) models must be robust to unexpected missing or noisy modalities during testing. In this paper, we propose to optimize for a joint generative-discriminative objective across multimodal data and labels. We introduce a model that factorizes representations into two sets of independent factors: multimodal discriminative and modality-specific generative factors. Multimodal discriminative factors are shared across all modalities and contain joint multimodal features required for discriminative tasks such as sentiment prediction. Modality-specific generative factors are unique for each modality and contain the information required for generating data. Experimental results show that our model is able to learn meaningful multimodal representations that achieve state-of-the-art or competitive performance on six multimodal datasets. Our model demonstrates flexible generative capabilities by conditioning on independent factors and can reconstruct missing modalities without significantly impacting performance. Lastly, we interpret our factorized representations to understand the interactions that influence multimodal learning. * equal contributions arXiv:1806.06176v2 [cs.LG] | LEARNING FACTORIZED MULTIMODAL REPRESENTATIONS |
d256105170 | Real-world data generation often involves complex inter-dependencies among instances, violating the IID-data hypothesis of standard learning paradigms and posing a challenge for uncovering the geometric structures for learning desired instance representations. To this end, we introduce an energy constrained diffusion model which encodes a batch of instances from a dataset into evolutionary states that progressively incorporate other instances' information by their interactions. The diffusion process is constrained by descent criteria w.r.t. a principled energy function that characterizes the global consistency of instance representations over latent structures. We provide rigorous theory that implies closed-form optimal estimates for the pairwise diffusion strength among arbitrary instance pairs, which gives rise to a new class of neural encoders, dubbed as DIFFORMER (diffusion-based Transformers), with two instantiations: a simple version with linear complexity for prohibitive instance numbers, and an advanced version for learning complex structures. Experiments highlight the wide applicability of our model as a general-purpose encoder backbone with superior performance in various tasks, such as node classification on large graphs, semi-supervised image/text classification, and spatial-temporal dynamics prediction. The codes are available at https://github.com/qitianwu/DIFFormer. | DIFFORMER: SCALABLE (GRAPH) TRANSFORMERS INDUCED BY ENERGY CONSTRAINED DIFFUSION |
d264439585 | Privacy amplification exploits randomness in data selection to provide tighter differential privacy (DP) guarantees.This analysis is key to DP-SGD's success in machine learning (ML), but, is not readily applicable to the newer state-of-the-art (SOTA) algorithms.This is because these algorithms, known as DP-FTRL, use the matrix mechanism to add correlated noise instead of independent noise as in DP-SGD.In this paper, we propose "MMCC", the first algorithm to analyze privacy amplification via sampling for any generic matrix mechanism.MMCC is nearly tight in that it approaches a lower bound as ε → 0. To analyze correlated outputs in MMCC, we prove that they can be analyzed as if they were independent, by conditioning them on prior outputs.Our "conditional composition theorem" has broad utility: we use it to show that the noise added to binary-tree-DP-FTRL can asymptotically match the noise added to DP-SGD with amplification.Our amplification algorithm also has practical empirical utility: we show it leads to significant improvement in the privacy-utility trade-offs for DP-FTRL algorithms on standard benchmarks. | Privacy Amplification for Matrix Mechanisms |
d12462234 | The capacity of a neural network to absorb information is limited by its number of parameters. Conditional computation, where parts of the network are active on a per-example basis, has been proposed in theory as a way of dramatically increasing model capacity without a proportional increase in computation. In practice, however, there are significant algorithmic and performance challenges. In this work, we address these challenges and finally realize the promise of conditional computation, achieving greater than 1000x improvements in model capacity with only minor losses in computational efficiency on modern GPU clusters. We introduce a Sparsely-Gated Mixture-of-Experts layer (MoE), consisting of up to thousands of feed-forward sub-networks. A trainable gating network determines a sparse combination of these experts to use for each example. We apply the MoE to the tasks of language modeling and machine translation, where model capacity is critical for absorbing the vast quantities of knowledge available in the training corpora. We present model architectures in which a MoE with up to 137 billion parameters is applied convolutionally between stacked LSTM layers. On large language modeling and machine translation benchmarks, these models achieve significantly better results than state-of-the-art at lower computational cost. * Equally major contributors † Work done as a member of the Google Brain Residency program (g.co/brainresidency) Under review as a conference paper at ICLR 2017 Figure 1: A Mixture of Experts (MoE) layer embedded within a recurrent language model. In this case, the sparse gating function selects two experts to perform computations. Their outputs are modulated by the outputs of the gating network.While these ideas are promising in theory, no work to date has yet demonstrated massive improvements in model capacity, training time, or model quality. We blame this on a combination of the following challenges:• Modern computing devices, especially GPUs, are much faster at arithmetic than at branching. Most of the works above recognize this and propose turning on/off large chunks of the network with each gating decision.• Large batch sizes are critical for performance, as they amortize the costs of parameter transfers and updates. Conditional computation reduces the batch sizes for the conditionally active chunks of the network.• Network bandwidth can be a bottleneck. A cluster of GPUs may have computational power thousands of times greater than the aggregate inter-device network bandwidth. To be computationally efficient, the relative computational versus network demands of an algorithm must exceed this ratio. Embedding layers, which can be seen as a form of conditional computation, are handicapped by this very problem. Since the embeddings generally need to be sent across the network, the number of (example, parameter) interactions is limited by network bandwidth instead of computational capacity.• Depending on the scheme, loss terms may be necessary to achieve the desired level of sparsity per-chunk and/or per example. Bengio et al.(2015)use three such terms. These issues can affect both model quality and load-balancing. diction with lstm. Neural Computation, 2000. Audrunas Gruslys, Rémi Munos, Ivo Danihelka, Marc Lanctot, and Alex Graves. Memory-efficient backpropagation through time. CoRR, abs/1606.03401, 2016. URL http://arxiv.org/ abs/1606.03401. Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. | OUTRAGEOUSLY LARGE NEURAL NETWORKS: THE SPARSELY-GATED MIXTURE-OF-EXPERTS LAYER |
d204207649 | Transferring knowledge across tasks to improve data-efficiency is one of the open key challenges in the field of global black-box optimization. Readily available algorithms are typically designed to be universal optimizers and, therefore, often suboptimal for specific tasks. We propose a novel transfer learning method to obtain customized optimizers within the well-established framework of Bayesian optimization, allowing our algorithm to utilize the proven generalization capabilities of Gaussian processes. Using reinforcement learning to meta-train an acquisition function (AF) on a set of related tasks, the proposed method learns to extract implicit structural information and to exploit it for improved data-efficiency. We present experiments on a simulation-to-real transfer task as well as on several synthetic functions and on two hyperparameter search problems. The results show that our algorithm (1) automatically identifies structural properties of objective functions from available source tasks or simulations, (2) performs favourably in settings with both scarse and abundant source data, and (3) falls back to the performance level of general AFs if no particular structure is present. | META-LEARNING ACQUISITION FUNCTIONS FOR TRANSFER LEARNING IN BAYESIAN OPTIMIZATION |
d263334596 | Molecular Representation Learning (MRL) has proven impactful in numerous biochemical applications such as drug discovery and enzyme design.While Graph Neural Networks (GNNs) are effective at learning molecular representations from a 2D molecular graph or a single 3D structure, existing works often overlook the flexible nature of molecules, which continuously interconvert across conformations via chemical bond rotations and minor vibrational perturbations.To better account for molecular flexibility, some recent works formulate MRL as an ensemble learning problem, focusing on explicitly learning from a set of conformer structures.However, most of these studies have limited datasets, tasks, and models.In this work, we introduce the first MoleculAR Conformer Ensemble Learning (MARCEL) benchmark to thoroughly evaluate the potential of learning on conformer ensembles and suggest promising research directions.MARCEL includes four datasets covering diverse molecule-and reaction-level properties of chemically diverse molecules including organocatalysts and transition-metal catalysts, extending beyond the scope of common GNN benchmarks that are confined to drug-like molecules.In addition, we conduct a comprehensive empirical study, which benchmarks representative 1D, 2D, and 3D molecular representation learning models, along with two strategies that explicitly incorporate conformer ensembles into 3D MRL models.Our findings reveal that direct learning from an accessible conformer space can improve performance on a variety of tasks and models. | Learning Over Molecular Conformer Ensembles: Datasets and Benchmarks |
d248986239 | Chain-of-thought prompting has demonstrated remarkable performance on various natural language reasoning tasks. However, it tends to perform poorly on tasks which requires solving problems harder than the exemplars shown in the prompts. To overcome this challenge of easy-to-hard generalization, we propose a novel prompting strategy, least-to-most prompting. The key idea in this strategy is to break down a complex problem into a series of simpler subproblems and then solve them in sequence. Solving each subproblem is facilitated by the answers to previously solved subproblems. Our experimental results on tasks related to symbolic manipulation, compositional generalization, and math reasoning reveal that least-to-most prompting is capable of generalizing to more difficult problems than those seen in the prompts. A notable finding is that when the GPT-3 code-davinci-002 model is used with least-to-most prompting, it can solve the compositional generalization benchmark SCAN in any split (including length split) with an accuracy of at least 99% using just 14 exemplars, compared to only 16% accuracy with chain-of-thought prompting. This is particularly noteworthy because neural-symbolic models in the literature that specialize in solving SCAN are trained on the entire training set containing over 15,000 examples. We have included prompts for all the tasks in the Appendix. * Corresponding to: dennyzhou@google.com Published as a conference paper at ICLR 2023 subproblems. Both stages are implemented by few-shot prompting, so that there is no training or finetuning in either stage. An example usage of least-to-most prompting is illustrated inFigure 1.The term least-to-most prompting is borrowed from educational psychology (Libby et al., 2008), where it is used to denote the technique of using a progressive sequence of prompts to help a student to learn a new skill. Here we apply this technique for teaching humans to teach language models. Empirical results on symbolic manipulation, compositional generalization, and math reasoning show that least-to-most prompting can indeed generalize to problems harder than those demonstrated.Figure 1: Least-to-most prompting solving a math word problem in two stages: (1) query the language model to decompose the problem into subproblems; (2) query the language model to sequentially solve the subproblems. The answer to the second subproblem is built on the answer to the first subproblem. The demonstration examples for each stage's prompt are omitted in this illustration.LEAST-TO-MOST PROMPTINGLeast-to-most prompting teaches language models how to solve a complex problem by decomposing it to a series of simpler subproblems. It consists of two sequential stages:1. Decomposition. The prompt in this stage contains constant examples that demonstrate the decomposition, followed by the specific question to be decomposed.2. Subproblem solving. The prompt in this stage consists of three parts: (1) constant examples demonstrating how subproblems are solved;(2) a potentially empty list of previously answered subquestions and generated solutions, and (3) the question to be answered next.In the example shown inFigure 1, the language model is first asked to decompose the original problem into subproblems. The prompt that is passed to the model consists of examples that illustrate how to decompose complex problems (which are not shown in the figure), followed by the specific problem to be decomposed (as shown in the figure). The language model figures out that the original problem can be solved via solving an intermediate problem "How long does each trip take?". | LEAST-TO-MOST PROMPTING ENABLES COMPLEX REASONING IN LARGE LANGUAGE MODELS |
d220265992 | Kernelized Stein discrepancy (KSD), though being extensively used in goodness-offit tests and model learning, suffers from the curse-of-dimensionality. We address this issue by proposing the sliced Stein discrepancy and its scalable and kernelized variants, which employs kernel-based test functions defined on the optimal onedimensional projections instead of the full input in high dimensions. When applied to goodness-of-fit tests, extensive experiments show the proposed discrepancy significantly outperforms KSD and various baselines in high dimensions. For model learning, we show its advantages by training an independent component analysis when compared with existing Stein discrepancy baselines. We further propose a novel particle inference method called sliced Stein variational gradient descent (S-SVGD) which alleviates the mode-collapse issue of SVGD in training variational autoencoders.Preprint. Under review. | Sliced Kernelized Stein Discrepancy |
d56895597 | Invariant and equivariant networks have been successfully used for learning images, sets, point clouds, and graphs. A basic challenge in developing such networks is finding the maximal collection of invariant and equivariant linear layers. Although this question is answered for the first three examples (for popular transformations, at-least), a full characterization of invariant and equivariant linear layers for graphs is not known. In this paper we provide a characterization of all permutation invariant and equivariant linear layers for (hyper-)graph data, and show that their dimension, in case of edge-value graph data, is 2 and 15, respectively. More generally, for graph data defined on k-tuples of nodes, the dimension is the k-th and 2k-th Bell numbers. Orthogonal bases for the layers are computed, including generalization to multigraph data. The constant number of basis elements and their characteristics allow successfully applying the networks to different size graphs. From the theoretical point of view, our results generalize and unify recent advancement in equivariant deep learning. In particular, we show that our model is capable of approximating any message passing neural network. Applying these new linear layers in a simple deep neural network framework is shown to achieve comparable results to state-of-the-art and to have better expressivity than previous invariant and equivariant bases. 1 A hyper-edge is an ordered subset of the nodes, V | INVARIANT AND EQUIVARIANT GRAPH NETWORKS |
d3544583 | The rapid adoption of machine learning has increased concerns about the privacy implications of machine learning models trained on sensitive data, such as medical records or other personal information. To address those concerns, one promising approach is Private Aggregation of Teacher Ensembles, or PATE, which transfers to a "student" model the knowledge of an ensemble of "teacher" models, with intuitive privacy provided by training teachers on disjoint data and strong privacy guaranteed by noisy aggregation of teachers' answers. However, PATE has so far been evaluated only on simple classification tasks like MNIST, leaving unclear its utility when applied to larger-scale learning tasks and real-world datasets. In this work, we show how PATE can scale to learning tasks with large numbers of output classes and uncurated, imbalanced training data with errors. For this, we introduce new noisy aggregation mechanisms for teacher ensembles that are more selective and add less noise, and prove their tighter differential-privacy guarantees. Our new mechanisms build on two insights: the chance of teacher consensus is increased by using more concentrated noise and, lacking consensus, no answer need be given to a student. The consensus answers used are more likely to be correct, offer better intuitive privacy, and incur lower-differential privacy cost. Our evaluation shows our mechanisms improve on the original PATE on all measures, and scale to larger tasks with both high utility and very strong privacy (ε < 1.0). | SCALABLE PRIVATE LEARNING WITH PATE |
d232170254 | Off-policy evaluation (OPE) is the task of estimating the expected reward of a given policy based on offline data previously collected under different policies. Therefore, OPE is a key step in applying reinforcement learning to real-world domains such as medical treatment, where interactive data collection is expensive or even unsafe. As the observed data tends to be noisy and limited, it is essential to provide rigorous uncertainty quantification, not just a point estimation, when applying OPE to make high stakes decisions. This work considers the problem of constructing non-asymptotic confidence intervals in infinite-horizon off-policy evaluation, which remains a challenging open question. We develop a practical algorithm through a primal-dual optimization-based approach, which leverages the kernel Bellman loss (KBL) of Feng et al. (2019) and a new martingale concentration inequality of KBL applicable to timedependent data with unknown mixing conditions. Our algorithm makes minimum assumptions on the data and the function class of the Q-function, and works for the behavior-agnostic settings where the data is collected under a mix of arbitrary unknown behavior policies. We present empirical results that clearly demonstrate the advantages of our approach over existing methods. * Equal contribution. This article is an extended version ofFeng et al. (2021)in ICLR 2021. | Non-asymptotic Confidence Intervals of Off-policy Evaluation: Primal and Dual Bounds |
d254685748 | Boundary conditions (BCs) are important groups of physics-enforced constraints that are necessary for solutions of Partial Differential Equations (PDEs) to satisfy at specific spatial locations. These constraints carry important physical meaning, and guarantee the existence and the uniqueness of the PDE solution. Current neural-network based approaches that aim to solve PDEs rely only on training data to help the model learn BCs implicitly. There is no guarantee of BC satisfaction by these models during evaluation. In this work, we propose Boundary enforcing Operator Network (BOON) that enables the BC satisfaction of neural operators by making structural changes to the operator kernel. We provide our refinement procedure, and demonstrate the satisfaction of physics-based BCs, e.g. Dirichlet, Neumann, and periodic by the solutions obtained by BOON. Numerical experiments based on multiple PDEs with a wide variety of applications indicate that the proposed approach ensures satisfaction of BCs, and leads to more accurate solutions over the entire domain. The proposed correction method exhibits a (2X-20X) improvement over a given operator model in relative L 2 error (0.000084 relative L 2 error for Burgers' equation). | Guiding continuous operator learning through Physics-based boundary constraints |
d254018206 | We present a novel neural network architecture using self-attention, the Wavefunction Transformer (Psiformer), which can be used as an approximation (or Ansatz) for solving the many-electron Schrödinger equation, the fundamental equation for quantum chemistry and material science. This equation can be solved from first principles, requiring no external training data. In recent years, deep neural networks like the FermiNet and PauliNet have been used to significantly improve the accuracy of these first-principle calculations, but they lack an attention-like mechanism for gating interactions between electrons. Here we show that the Psiformer can be used as a drop-in replacement for these other neural networks, often dramatically improving the accuracy of the calculations. On larger molecules especially, the ground state energy can be improved by dozens of kcal/mol, a qualitative leap over previous methods. This demonstrates that self-attention networks can learn complex quantum mechanical correlations between electrons, and are a promising route to reaching unprecedented accuracy in chemical calculations on larger systems. | A Self-Attention Ansatz for Ab-initio Quantum Chemistry |
d53415112 | Reinforcement learning agents need exploratory behaviors to escape from local optima. These behaviors may include both immediate dithering perturbation and temporally consistent exploration. To achieve these, a stochastic policy model that is inherently consistent through a period of time is in desire, especially for tasks with either sparse rewards or long term information. In this work, we introduce a novel on-policy temporally consistent exploration strategy -Neural Adaptive Dropout Policy Exploration (NADPEx) -for deep reinforcement learning agents. Modeled as a global random variable for conditional distribution, dropout is incorporated to reinforcement learning policies, equipping them with inherent temporal consistency, even when the reward signals are sparse. Two factors, gradients' alignment with the objective and KL constraint in policy space, are discussed to guarantee NADPEx policy's stable improvement. Our experiments demonstrate that NADPEx solves tasks with sparse reward while naive exploration and parameter noise fail. It yields as well or even faster convergence in the standard mujoco benchmark for continuous control.Published as a conference paper at ICLR 2019 unadjustable, and the duplicated modules do not cooperate with each other. Directly optimizing the mean of distribution of all parameters, Plappert et al. (2017) adjusts the stochasticity to the learning progress heristically. Besides, as all sampled parameters need to be stored in on-policy exploration, it is not directly salable to large neural networks, where the trade-off between large memory for multiple networks and high variance with sinlge network is non-trivial to make. This work proposes Neural Adaptive Dropout Policy Exploration (NADPEx) as a simple, scalable and improvement-guaranteed method for on-policy temporally consistent exploration, which generalizes Osband et al.(2016)and Plappert et al. (2017). Policy stochasticity is disentangled into two time scales, step-wise action noise and episode-wise stochasticity modeled with a distribution of subnetworks. With one single subnetwork in an episode, it achieves inherent temporal consistency with only a little computational overhead and no crafted reward, even in environments with sparse rewards. And with a set of subnetworks in one sampling batch, agents experience diverse behavioral patterns. This sampling of subnetworks is executed through dropout (Hinton et al., 2012; Srivastava et al., 2014), which encourages composability of constituents and facilitates the emergence of diverse maneuvers. As dropout could be excerted on connections, neurons, modules and paths, NADPEx naturally extends to neural networks with various sizes, exploiting modularity at various levels. To align separately parametrized stochasticity to each other, this sampling is made differentiable for all possible dropout candidates. We further discuss the effect of a first-order KL regularizer on dropout policies to improve stability and guarantee policy improvement. Our experiments demonstrate that NAPDEx solves challenging exploration tasks with sparse rewards while achieving as efficient or even faster convergence in the standard mujoco benchmark for state-of-theart PPO agents, which we believe can be generalized to any deep reinforcement learning agents. | NADPEX: AN ON-POLICY TEMPORALLY CONSISTENT EXPLORATION METHOD FOR DEEP REINFORCEMENT LEARNING |
d246430145 | Learning multiple tasks sequentially without forgetting previous knowledge, called Continual Learning (CL), remains a long-standing challenge for neural networks. Most existing methods rely on additional network capacity or data replay. In contrast, we introduce a novel approach which we refer to as Recursive Gradient Optimization (RGO). RGO is composed of an iteratively updated optimizer that modifies the gradient to minimize forgetting without data replay and a virtual Feature Encoding Layer (FEL) that represents different network structures with only task descriptors. Experiments demonstrate that RGO has significantly better performance on popular continual classification benchmarks when compared to the baselines and achieves new state-of-the-art performance on 20-split-CIFAR100 (82.22%) and 20-split-miniImageNet (72.63%). With higher average accuracy than Single-Task Learning (STL), this method is flexible and reliable to provide continual learning capabilities for learning models that rely on gradient descent. | CONTINUAL LEARNING WITH RECURSIVE GRADIENT OPTIMIZATION |
d247011095 | Human intervention is an effective way to inject human knowledge into the loop of reinforcement learning, bringing fast learning and training safety. But given the very limited budget of human intervention, it is challenging to design when and how human expert interacts with the learning agent in the training. In this work, we develop a novel human-in-the-loop learning method called Human-AI Copilot Optimization (HACO). To allow the agent's sufficient exploration in the risky environments while ensuring the training safety, the human expert can take over the control and demonstrate to the agent how to avoid probably dangerous situations or trivial behaviors. The proposed HACO then effectively utilizes the data collected both from the trial-and-error exploration and human's partial demonstration to train a high-performing agent. HACO extracts proxy state-action values from partial human demonstration and optimizes the agent to improve the proxy values while reducing the human interventions. No environmental reward is required in HACO. The experiments show that HACO achieves a substantially high sample efficiency in the safe driving benchmark. It can train agents to drive in unseen traffic scenes with a handful of human intervention budget and achieve high safety and generalizability, outperforming both reinforcement learning and imitation learning baselines with a large margin. Code and demo videos are available at: https://decisionforce.github.io/HACO/. * Quanyi Li and Zhenghao Peng contribute equally to this work. | EFFICIENT LEARNING OF SAFE DRIVING POLICY VIA HUMAN-AI COPILOT OPTIMIZATION |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.