id
stringlengths
9
16
title
stringlengths
4
278
categories
stringlengths
5
104
abstract
stringlengths
6
4.09k
2501.11924
Make Full Use of Testing Information: An Integrated Accelerated Testing and Evaluation Method for Autonomous Driving Systems
cs.AI
Testing and evaluation is an important step before the large-scale application of the autonomous driving systems (ADSs). Based on the three level of scenario abstraction theory, a testing can be performed within a logical scenario, followed by an evaluation stage which is inputted with the testing results of each concrete scenario generated from the logical parameter space. During the above process, abundant testing information is produced which is beneficial for comprehensive and accurate evaluations. To make full use of testing information, this paper proposes an Integrated accelerated Testing and Evaluation Method (ITEM). Based on a Monte Carlo Tree Search (MCTS) paradigm and a dual surrogates testing framework proposed in our previous work, this paper applies the intermediate information (i.e., the tree structure, including the affiliation of each historical sampled point with the subspaces and the parent-child relationship between subspaces) generated during the testing stage into the evaluation stage to achieve accurate hazardous domain identification. Moreover, to better serve this purpose, the UCB calculation method is improved to allow the search algorithm to focus more on the hazardous domain boundaries. Further, a stopping condition is constructed based on the convergence of the search algorithm. Ablation and comparative experiments are then conducted to verify the effectiveness of the improvements and the superiority of the proposed method. The experimental results show that ITEM could well identify the hazardous domains in both low- and high-dimensional cases, regardless of the shape of the hazardous domains, indicating its generality and potential for the safety evaluation of ADSs.
2501.11926
Multi-Modal Variable-Rate CSI Reconstruction for FDD Massive MIMO Systems
cs.IT eess.SP math.IT
In frequency division duplex (FDD) systems, acquiring channel state information (CSI) at the base station (BS) traditionally relies on limited feedback from mobile terminals (MTs). However, the accuracy of channel reconstruction from feedback CSI is inherently constrained by the rate-distortion trade-off. To overcome this limitation, we propose a multi-modal channel reconstruction framework that leverages auxiliary data, such as RGB images or uplink CSI, collected at the BS. By integrating contextual information from these modalities, the framework mitigates CSI distortions caused by noise, compression, and quantization. At its core, the framework utilizes an autoencoder network capable of generating variable-length CSI, tailored for rate-adaptive multi-modal channel reconstruction. By augmenting the foundational autoencoder network using a transfer learning-based multi-modal fusion strategy, we enable accurate channel reconstruction in both single-modal and multi-modal scenarios. To train and evaluate the network under diverse and realistic wireless conditions, we construct a synthetic dataset that pairs wireless channel data with sensor data through 3D modeling and ray tracing. Simulation results demonstrate that the proposed framework achieves near-optimal beamforming gains in 5G New Radio (5G NR)-compliant scenarios, highlighting the potential of sensor data integration to improve CSI reconstruction accuracy.
2501.11927
A Lightweight and Interpretable Deepfakes Detection Framework
cs.CV cs.AI
The recent realistic creation and dissemination of so-called deepfakes poses a serious threat to social life, civil rest, and law. Celebrity defaming, election manipulation, and deepfakes as evidence in court of law are few potential consequences of deepfakes. The availability of open source trained models based on modern frameworks such as PyTorch or TensorFlow, video manipulations Apps such as FaceApp and REFACE, and economical computing infrastructure has easen the creation of deepfakes. Most of the existing detectors focus on detecting either face-swap, lip-sync, or puppet master deepfakes, but a unified framework to detect all three types of deepfakes is hardly explored. This paper presents a unified framework that exploits the power of proposed feature fusion of hybrid facial landmarks and our novel heart rate features for detection of all types of deepfakes. We propose novel heart rate features and fused them with the facial landmark features to better extract the facial artifacts of fake videos and natural variations available in the original videos. We used these features to train a light-weight XGBoost to classify between the deepfake and bonafide videos. We evaluated the performance of our framework on the world leaders dataset (WLDR) that contains all types of deepfakes. Experimental results illustrate that the proposed framework offers superior detection performance over the comparative deepfakes detection methods. Performance comparison of our framework against the LSTM-FCN, a candidate of deep learning model, shows that proposed model achieves similar results, however, it is more interpretable.
2501.11929
ALoFTRAG: Automatic Local Fine Tuning for Retrieval Augmented Generation
cs.LG
Retrieval Augmented Generation (RAG) systems have been shown to improve the accuracy of Large Language Model (LLM) outputs. However, these models can often achieve low accuracy when applied to new data domains. We introduce the Automatic Local Fine Tuning of Retrieval Augmented Generation models (ALoFTRAG) framework, designed to improve the accuracy of RAG systems on a given domain by training LLMs without manually labeled data or using larger teacher models. By generating and filtering synthetic training data and performing LoRA fine-tuning, ALoFTRAG improves citation and answer accuracy across 20 datasets in 26 languages by, on average, 8.3% and 3.0% respectively. Our results demonstrate that ALoFTRAG offers a practical, cost-effective, and data-secure solution for improving RAG accuracy, making it particularly applicable to sensitive domains such as healthcare and finance.
2501.11930
Nocturnal eye inspired liquid to gas phase change soft actuator with Laser-Induced-Graphene: enhanced environmental light harvesting and photothermal conversion
cs.RO
Robotic systems' mobility is constrained by power sources and wiring. While pneumatic actuators remain tethered to air supplies, we developed a new actuator utilizing light energy. Inspired by nocturnal animals' eyes, we designed a bilayer soft actuator incorporating Laser-Induced Graphene (LIG) on the inner surface of a silicone layer. This design maintains silicone's transparency and flexibility while achieving 54% faster response time compared to conventional actuators through enhanced photothermal conversion.
2501.11931
Construction of Simultaneously Good Polar Codes and Polar Lattices
cs.IT math.IT
In this work, we investigate the simultaneous goodness of polar codes and polar lattices. The simultaneous goodness of a lattice or a code means that it is optimal for both channel coding and source coding simultaneously. The existence of such kind of lattices was proven by using random lattice ensembles. Our work provides an explicit construction based on the polarization technique.
2501.11935
Web vs. LLMs: An Empirical Study of Learning Behaviors of CS2 Students
cs.HC cs.AI
LLMs such as ChatGPT have been widely adopted by students in higher education as tools for learning programming and related concepts. However, it remains unclear how effective students are and what strategies students use while learning with LLMs. Since the majority of students' experiences in online self-learning have come through using search engines such as Google, evaluating AI tools in this context can help us address these gaps. In this mixed methods research, we conducted an exploratory within-subjects study to understand how CS2 students learn programming concepts using both LLMs as well as traditional online methods such as educational websites and videos to examine how students approach learning within and across both scenarios. We discovered that students found it easier to learn a more difficult concept using traditional methods than using ChatGPT. We also found that students ask fewer follow-ups and use more keyword-based queries for search engines while their prompts to LLMs tend to explicitly ask for information.
2501.11937
MeshONet: A Generalizable and Efficient Operator Learning Method for Structured Mesh Generation
cs.LG cs.AI
Mesh generation plays a crucial role in scientific computing. Traditional mesh generation methods, such as TFI and PDE-based methods, often struggle to achieve a balance between efficiency and mesh quality. To address this challenge, physics-informed intelligent learning methods have recently emerged, significantly improving generation efficiency while maintaining high mesh quality. However, physics-informed methods fail to generalize when applied to previously unseen geometries, as even small changes in the boundary shape necessitate burdensome retraining to adapt to new geometric variations. In this paper, we introduce MeshONet, the first generalizable intelligent learning method for structured mesh generation. The method transforms the mesh generation task into an operator learning problem with multiple input and solution functions. To effectively overcome the multivariable mapping restriction of operator learning methods, we propose a dual-branch, shared-trunk architecture to approximate the mapping between function spaces based on input-output pairs. Experimental results show that MeshONet achieves a speedup of up to four orders of magnitude in generation efficiency over traditional methods. It also enables generalization to different geometries without retraining, greatly enhancing the practicality of intelligent methods.
2501.11938
Navigating Robot Swarm Through a Virtual Tube with Flow-Adaptive Distribution Control
cs.RO cs.SY eess.SY
With the rapid development of robot swarm technology and its diverse applications, navigating robot swarms through complex environments has emerged as a critical research direction. To ensure safe navigation and avoid potential collisions with obstacles, the concept of virtual tubes has been introduced to define safe and navigable regions. However, current control methods in virtual tubes face the congestion issues, particularly in narrow virtual tubes with low throughput. To address these challenges, we first originally introduce the concepts of virtual tube area and flow capacity, and develop an new evolution model for the spatial density function. Next, we propose a novel control method that combines a modified artificial potential field (APF) for swarm navigation and density feedback control for distribution regulation, under which a saturated velocity command is designed. Then, we generate a global velocity field that not only ensures collision-free navigation through the virtual tube, but also achieves locally input-to-state stability (LISS) for density tracking errors, both of which are rigorously proven. Finally, numerical simulations and realistic applications validate the effectiveness and advantages of the proposed method in managing robot swarms within narrow virtual tubes.
2501.11945
Learning to Hop for a Single-Legged Robot with Parallel Mechanism
cs.RO
This work presents the application of reinforcement learning to improve the performance of a highly dynamic hopping system with a parallel mechanism. Unlike serial mechanisms, parallel mechanisms can not be accurately simulated due to the complexity of their kinematic constraints and closed-loop structures. Besides, learning to hop suffers from prolonged aerial phase and the sparse nature of the rewards. To address them, we propose a learning framework to encode long-history feedback to account for the under-actuation brought by the prolonged aerial phase. In the proposed framework, we also introduce a simplified serial configuration for the parallel design to avoid directly simulating parallel structure during the training. A torque-level conversion is designed to deal with the parallel-serial conversion to handle the sim-to-real issue. Simulation and hardware experiments have been conducted to validate this framework.
2501.11949
GLAM: Global-Local Variation Awareness in Mamba-based World Model
cs.LG
Mimicking the real interaction trajectory in the inference of the world model has been shown to improve the sample efficiency of model-based reinforcement learning (MBRL) algorithms. Many methods directly use known state sequences for reasoning. However, this approach fails to enhance the quality of reasoning by capturing the subtle variation between states. Much like how humans infer trends in event development from this variation, in this work, we introduce Global-Local variation Awareness Mamba-based world model (GLAM) that improves reasoning quality by perceiving and predicting variation between states. GLAM comprises two Mambabased parallel reasoning modules, GMamba and LMamba, which focus on perceiving variation from global and local perspectives, respectively, during the reasoning process. GMamba focuses on identifying patterns of variation between states in the input sequence and leverages these patterns to enhance the prediction of future state variation. LMamba emphasizes reasoning about unknown information, such as rewards, termination signals, and visual representations, by perceiving variation in adjacent states. By integrating the strengths of the two modules, GLAM accounts for highervalue variation in environmental changes, providing the agent with more efficient imagination-based training. We demonstrate that our method outperforms existing methods in normalized human scores on the Atari 100k benchmark.
2501.11951
HERITAGE: An End-to-End Web Platform for Processing Korean Historical Documents in Hanja
cs.CL
While Korean historical documents are invaluable cultural heritage, understanding those documents requires in-depth Hanja expertise. Hanja is an ancient language used in Korea before the 20th century, whose characters were borrowed from old Chinese but had evolved in Korea for centuries. Modern Koreans and Chinese cannot understand Korean historical documents without substantial additional help, and while previous efforts have produced some Korean and English translations, this requires in-depth expertise, and so most of the documents are not translated into any modern language. To address this gap, we present HERITAGE, the first open-source Hanja NLP toolkit to assist in understanding and translating the unexplored Korean historical documents written in Hanja. HERITAGE is a web-based platform providing model predictions of three critical tasks in historical document understanding via Hanja language models: punctuation restoration, named entity recognition, and machine translation (MT). HERITAGE also provides an interactive glossary, which provides the character-level reading of the Hanja characters in modern Korean, as well as character-level English definition. HERITAGE serves two purposes. First, anyone interested in these documents can get a general understanding from the model predictions and the interactive glossary, especially MT outputs in Korean and English. Second, since the model outputs are not perfect, Hanja experts can revise them to produce better annotations and translations. This would boost the translation efficiency and potentially lead to most of the historical documents being translated into modern languages, lowering the barrier on unexplored Korean historical documents.
2501.11953
Proverbs Run in Pairs: Evaluating Proverb Translation Capability of Large Language Model
cs.CL
Despite achieving remarkable performance, machine translation (MT) research remains underexplored in terms of translating cultural elements in languages, such as idioms, proverbs, and colloquial expressions. This paper investigates the capability of state-of-the-art neural machine translation (NMT) and large language models (LLMs) in translating proverbs, which are deeply rooted in cultural contexts. We construct a translation dataset of standalone proverbs and proverbs in conversation for four language pairs. Our experiments show that the studied models can achieve good translation between languages with similar cultural backgrounds, and LLMs generally outperform NMT models in proverb translation. Furthermore, we find that current automatic evaluation metrics such as BLEU, CHRF++ and COMET are inadequate for reliably assessing the quality of proverb translation, highlighting the need for more culturally aware evaluation metrics.
2501.11959
Noise-Resilient Point-wise Anomaly Detection in Time Series Using Weak Segment Labels
cs.LG
Detecting anomalies in temporal data has gained significant attention across various real-world applications, aiming to identify unusual events and mitigate potential hazards. In practice, situations often involve a mix of segment-level labels (detected abnormal events with segments of time points) and unlabeled data (undetected events), while the ideal algorithmic outcome should be point-level predictions. Therefore, the huge label information gap between training data and targets makes the task challenging. In this study, we formulate the above imperfect information as noisy labels and propose NRdetector, a noise-resilient framework that incorporates confidence-based sample selection, robust segment-level learning, and data-centric point-level detection for multivariate time series anomaly detection. Particularly, to bridge the information gap between noisy segment-level labels and missing point-level labels, we develop a novel loss function that can effectively mitigate the label noise and consider the temporal features. It encourages the smoothness of consecutive points and the separability of points from segments with different labels. Extensive experiments on real-world multivariate time series datasets with 11 different evaluation metrics demonstrate that NRdetector consistently achieves robust results across multiple real-world datasets, outperforming various baselines adapted to operate in our setting.
2501.11960
TAD-Bench: A Comprehensive Benchmark for Embedding-Based Text Anomaly Detection
cs.CL cs.AI
Text anomaly detection is crucial for identifying spam, misinformation, and offensive language in natural language processing tasks. Despite the growing adoption of embedding-based methods, their effectiveness and generalizability across diverse application scenarios remain under-explored. To address this, we present TAD-Bench, a comprehensive benchmark designed to systematically evaluate embedding-based approaches for text anomaly detection. TAD-Bench integrates multiple datasets spanning different domains, combining state-of-the-art embeddings from large language models with a variety of anomaly detection algorithms. Through extensive experiments, we analyze the interplay between embeddings and detection methods, uncovering their strengths, weaknesses, and applicability to different tasks. These findings offer new perspectives on building more robust, efficient, and generalizable anomaly detection systems for real-world applications.
2501.11963
A Contrastive Framework with User, Item and Review Alignment for Recommendation
cs.IR
Learning effective latent representations for users and items is the cornerstone of recommender systems. Traditional approaches rely on user-item interaction data to map users and items into a shared latent space, but the sparsity of interactions often poses challenges. While leveraging user reviews could mitigate this sparsity, existing review-aware recommendation models often exhibit two key limitations. First, they typically rely on reviews as additional features, but reviews are not universal, with many users and items lacking them. Second, such approaches do not integrate reviews into the user-item space, leading to potential divergence or inconsistency among user, item, and review representations. To overcome these limitations, our work introduces a Review-centric Contrastive Alignment Framework for Recommendation (ReCAFR), which incorporates reviews into the core learning process, ensuring alignment among user, item, and review representations within a unified space. Specifically, we leverage two self-supervised contrastive strategies that not only exploit review-based augmentation to alleviate sparsity, but also align the tripartite representations to enhance robustness. Empirical studies on public benchmark datasets demonstrate the effectiveness and robustness of ReCAFR.
2501.11967
A Hybrid Attention Framework for Fake News Detection with Large Language Models
cs.CL
With the rapid growth of online information, the spread of fake news has become a serious social challenge. In this study, we propose a novel detection framework based on Large Language Models (LLMs) to identify and classify fake news by integrating textual statistical features and deep semantic features. Our approach utilizes the contextual understanding capability of the large language model for text analysis and introduces a hybrid attention mechanism to focus on feature combinations that are particularly important for fake news identification. Extensive experiments on the WELFake news dataset show that our model significantly outperforms existing methods, with a 1.5\% improvement in F1 score. In addition, we assess the interpretability of the model through attention heat maps and SHAP values, providing actionable insights for content review strategies. Our framework provides a scalable and efficient solution to deal with the spread of fake news and helps build a more reliable online information ecosystem.
2501.11968
Bridging Visualization and Optimization: Multimodal Large Language Models on Graph-Structured Combinatorial Optimization
cs.AI cs.LG
Graph-structured combinatorial challenges are inherently difficult due to their nonlinear and intricate nature, often rendering traditional computational methods ineffective or expensive. However, these challenges can be more naturally tackled by humans through visual representations that harness our innate ability for spatial reasoning. In this study, we propose transforming graphs into images to preserve their higher-order structural features accurately, revolutionizing the representation used in solving graph-structured combinatorial tasks. This approach allows machines to emulate human-like processing in addressing complex combinatorial challenges. By combining the innovative paradigm powered by multimodal large language models (MLLMs) with simple search techniques, we aim to develop a novel and effective framework for tackling such problems. Our investigation into MLLMs spanned a variety of graph-based tasks, from combinatorial problems like influence maximization to sequential decision-making in network dismantling, as well as addressing six fundamental graph-related issues. Our findings demonstrate that MLLMs exhibit exceptional spatial intelligence and a distinctive capability for handling these problems, significantly advancing the potential for machines to comprehend and analyze graph-structured data with a depth and intuition akin to human cognition. These results also imply that integrating MLLMs with simple optimization strategies could form a novel and efficient approach for navigating graph-structured combinatorial challenges without complex derivations, computationally demanding training and fine-tuning.
2501.11971
SMamba: Sparse Mamba for Event-based Object Detection
cs.CV
Transformer-based methods have achieved remarkable performance in event-based object detection, owing to the global modeling ability. However, they neglect the influence of non-event and noisy regions and process them uniformly, leading to high computational overhead. To mitigate computation cost, some researchers propose window attention based sparsification strategies to discard unimportant regions, which sacrifices the global modeling ability and results in suboptimal performance. To achieve better trade-off between accuracy and efficiency, we propose Sparse Mamba (SMamba), which performs adaptive sparsification to reduce computational effort while maintaining global modeling capability. Specifically, a Spatio-Temporal Continuity Assessment module is proposed to measure the information content of tokens and discard uninformative ones by leveraging the spatiotemporal distribution differences between activity and noise events. Based on the assessment results, an Information-Prioritized Local Scan strategy is designed to shorten the scan distance between high-information tokens, facilitating interactions among them in the spatial dimension. Furthermore, to extend the global interaction from 2D space to 3D representations, a Global Channel Interaction module is proposed to aggregate channel information from a global spatial perspective. Results on three datasets (Gen1, 1Mpx, and eTram) demonstrate that our model outperforms other methods in both performance and efficiency.
2501.11972
"FRAME: Forward Recursive Adaptive Model Extraction -- A Technique for Advance Feature Selection"
cs.LG
Feature selection is a crucial preprocessing step in machine learning, impacting model performance, interpretability, and computational efficiency. This study introduces a novel hybrid approach, the Forward Recursive Adaptive Model Extraction Technique (FRAME), which combines Forward Selection and Recursive Feature Elimination (RFE) to enhance feature selection across diverse datasets. FRAME integrates the strengths of both methods, balancing exploration and exploitation of features to optimize selection. A comprehensive evaluation of FRAME was conducted against traditional methods such as SelectKBest and Lasso Regression, using high-dimensional, noisy, and heterogeneous datasets. The results demonstrate that FRAME consistently delivers superior predictive performance based on downstream machine learning evaluation metrics. It effectively reduces dimensionality while maintaining robust model performance, making it particularly valuable for applications requiring interpretable and accurate predictions, such as biomedical diagnostics. This study highlights the importance of assessing feature selection methods across varied datasets to ensure their robustness and generalizability. The findings suggest that FRAME has significant potential for further enhancement, particularly through integration with deep learning architectures for adaptive and real-time feature selection in dynamic environments. By advancing feature selection methodologies, FRAME offers a practical and effective solution to improve machine learning applications across multiple domains.
2501.11977
Leveraging Graph Structures and Large Language Models for End-to-End Synthetic Task-Oriented Dialogues
cs.CL cs.AI
Training task-oriented dialogue systems is both costly and time-consuming, due to the need for high-quality datasets encompassing diverse intents. Traditional methods depend on extensive human annotation, while recent advancements leverage large language models (LLMs) to generate synthetic data. However, these approaches often require custom prompts or code, limiting accessibility for non-technical users. We introduce GraphTOD, an end-to-end framework that simplifies the generation of task-oriented dialogues. Users can create dialogues by specifying transition graphs in JSON format. Our evaluation demonstrates that GraphTOD generates high-quality dialogues across various domains, significantly lowering the cost and complexity of dataset creation.
2501.11978
Weight Distribution of the Weighted Coordinates Poset Block Space and Singleton Bound
cs.IT math.CO math.IT
In this paper, we determine the complete weight distribution of the space $ \mathbb{F}_q^N $ endowed by the weighted coordinates poset block metric ($(P,w,\pi)$-metric), also known as the $(P,w,\pi)$-space, thereby obtaining it for $(P,w)$-space, $(P,\pi)$-space, $\pi$-space, and $P$-space as special cases. Further, when $P$ is a chain, the resulting space is called as Niederreiter-Rosenbloom-Tsfasman (NRT) weighted block space and when $P$ is hierarchical, the resulting space is called as weighted coordinates hierarchical poset block space. The complete weight distribution of both the spaces are deduced from the main result. Moreover, we define an $I$-ball for an ideal $I$ in $P$ and study the characteristics of it in $(P,w,\pi)$-space. We investigate the relationship between the $I$-perfect codes and $t$-perfect codes in $(P,w,\pi)$-space. Given an ideal $I$, we investigate how the maximum distance separability (MDS) is related with $I$-perfect codes and $t$-perfect codes in $(P,w,\pi)$-space. Duality theorem is derived for an MDS $(P,w,\pi)$-code when all the blocks are of same length. Finally, the distribution of codewords among $r$-balls is analyzed in the case of chain poset, when all the blocks are of same length.
2501.11979
Linear Feedback Control Systems for Iterative Prompt Optimization in Large Language Models
cs.LG
Large Language Models (LLMs) have revolutionized various applications by generating outputs based on given prompts. However, achieving the desired output requires iterative prompt refinement. This paper presents a novel approach that draws parallels between the iterative prompt optimization process in LLMs and feedback control systems. We iteratively refine the prompt by treating the deviation between the LLM output and the desired result as an error term until the output criteria are met. This process is akin to a feedback control system, where the LLM, despite being non-linear and non-deterministic, is managed using principles from linear feedback control systems. We explore the application of different types of controllers within this framework, providing a mathematical foundation for integrating linear feedback control mechanisms with LLMs.
2501.11980
A note on the sample complexity of multi-target detection
eess.SP cs.IT math.IT
This work studies the sample complexity of the multi-target detection (MTD) problem, which involves recovering a signal from a noisy measurement containing multiple instances of a target signal in unknown locations, each transformed by a random group element. This problem is primarily motivated by single-particle cryo-electron microscopy (cryo-EM), a groundbreaking technology for determining the structures of biological molecules. We establish upper and lower bounds for various MTD models in the high-noise regime as a function of the group, the distribution over the group, and the arrangement of signal occurrences within the measurement. The lower bounds are established through a reduction to the related multi-reference alignment problem, while the upper bounds are derived from explicit recovery algorithms utilizing autocorrelation analysis. These findings provide fundamental insights into estimation limits in noisy environments and lay the groundwork for extending this analysis to more complex applications, such as cryo-EM.
2501.11992
Survey on Hand Gesture Recognition from Visual Input
cs.CV cs.AI
Hand gesture recognition has become an important research area, driven by the growing demand for human-computer interaction in fields such as sign language recognition, virtual and augmented reality, and robotics. Despite the rapid growth of the field, there are few surveys that comprehensively cover recent research developments, available solutions, and benchmark datasets. This survey addresses this gap by examining the latest advancements in hand gesture and 3D hand pose recognition from various types of camera input data including RGB images, depth images, and videos from monocular or multiview cameras, examining the differing methodological requirements of each approach. Furthermore, an overview of widely used datasets is provided, detailing their main characteristics and application domains. Finally, open challenges such as achieving robust recognition in real-world environments, handling occlusions, ensuring generalization across diverse users, and addressing computational efficiency for real-time applications are highlighted to guide future research directions. By synthesizing the objectives, methodologies, and applications of recent studies, this survey offers valuable insights into current trends, challenges, and opportunities for future research in human hand gesture recognition.
2501.11993
Subcode Ensemble Decoding of Linear Block Codes
cs.IT math.IT
Low-density parity-check (LDPC) codes together with belief propagation (BP) decoding yield exceptional error correction capabilities in the large block length regime. Yet, there remains a gap between BP decoding and maximum likelihood decoding for short block length LDPC codes. In this context, ensemble decoding schemes yield both reduced latency and good error rates. In this paper, we propose subcode ensemble decoding (SCED), which employs an ensemble of decodings on different subcodes of the code. To ensure that all codewords are decodable, we use the concept of linear coverings and explore approaches for sampling suitable ensembles for short block length LDPC codes. Monte-Carlo simulations conducted for three LDPC codes demonstrate that SCED improves decoding performance compared to stand-alone decoding and automorphism ensemble decoding. In particular, in contrast to existing schemes, e.g., multiple bases belief propagation and automorphism ensemble decoding, SCED does not require the NP-complete search for low-weight dual codewords or knowledge of the automorphism group of the code, which is often unknown.
2501.12005
A note on the relations between mixture models, maximum-likelihood and entropic optimal transport
stat.ML cs.LG
This note aims to demonstrate that performing maximum-likelihood estimation for a mixture model is equivalent to minimizing over the parameters an optimal transport problem with entropic regularization. The objective is pedagogical: we seek to present this already known result in a concise and hopefully simple manner. We give an illustration with Gaussian mixture models by showing that the standard EM algorithm is a specific block-coordinate descent on an optimal transport loss.
2501.12009
Ratio Attack on G+G Convoluted Gaussian Signature
cs.CR cs.IT math.IT
A lattice-based signature, called G+G convoluted Gaussian signature was proposed in ASIACRYPT 2023 and was proved secure in the quantum random oracle model. In this paper, we propose a ratio attack on the G+G convoluted Gaussian signature to recover the secret key. The attack exploits the fact, proved in this paper, that the secret key can be obtained from the expected value of the ratio of signatures which follows a truncated Cauchy distribution. Moreover, we also compute the number of signatures required to successfully recover the secret key. Furthermore, we simulate the ratio attack in Sagemath with a few different parameters as a proof-of-concept of the ratio attack.
2501.12011
Reference-free Evaluation Metrics for Text Generation: A Survey
cs.CL
A number of automatic evaluation metrics have been proposed for natural language generation systems. The most common approach to automatic evaluation is the use of a reference-based metric that compares the model's output with gold-standard references written by humans. However, it is expensive to create such references, and for some tasks, such as response generation in dialogue, creating references is not a simple matter. Therefore, various reference-free metrics have been developed in recent years. In this survey, which intends to cover the full breadth of all NLG tasks, we investigate the most commonly used approaches, their application, and their other uses beyond evaluating models. The survey concludes by highlighting some promising directions for future research.
2501.12012
TabularARGN: A Flexible and Efficient Auto-Regressive Framework for Generating High-Fidelity Synthetic Data
cs.LG
Synthetic data generation for tabular datasets must balance fidelity, efficiency, and versatility to meet the demands of real-world applications. We introduce the Tabular Auto-Regressive Generative Network (TabularARGN), a flexible framework designed to handle mixed-type, multivariate, and sequential datasets. By training on all possible conditional probabilities, TabularARGN supports advanced features such as fairness-aware generation, imputation, and conditional generation on any subset of columns. The framework achieves state-of-the-art synthetic data quality while significantly reducing training and inference times, making it ideal for large-scale datasets with diverse structures. Evaluated across established benchmarks, including realistic datasets with complex relationships, TabularARGN demonstrates its capability to synthesize high-quality data efficiently. By unifying flexibility and performance, this framework paves the way for practical synthetic data generation across industries.
2501.12015
Full Proportional Justified Representation
cs.GT cs.AI
In multiwinner approval voting, forming a committee that proportionally represents voters' approval ballots is an essential task. The notion of justified representation (JR) demands that any large "cohesive" group of voters should be proportionally "represented". The "cohesiveness" is defined in different ways; two common ways are the following: (C1) demands that the group unanimously approves a set of candidates proportional to its size, while (C2) requires each member to approve at least a fixed fraction of such a set. Similarly, "representation" have been considered in different ways: (R1) the coalition's collective utility from the winning set exceeds that of any proportionally sized alternative, and (R2) for any proportionally sized alternative, at least one member of the coalition derives less utility from it than from the winning set. Three of the four possible combinations have been extensively studied: (C1)-(R1) defines Proportional Justified Representation (PJR), (C1)-(R2) defines Extended Justified Representation (EJR), (C2)-(R2) defines Full Justified Representation (FJR). All three have merits, but also drawbacks. PJR is the weakest notion, and perhaps not sufficiently demanding; EJR may not be compatible with perfect representation; and it is open whether a committee satisfying FJR can be found efficiently. We study the combination (C2)-(R1), which we call Full Proportional Justified Representation (FPJR). We investigate FPJR's properties and find that it shares PJR's advantages over EJR: several proportionality axioms (e.g. priceability, perfect representation) imply FPJR and PJR but not EJR. We also find that efficient rules like the greedy Monroe rule and the method of equal shares satisfy FPJR, matching a key advantage of EJR over FJR. However, the Proportional Approval Voting (PAV) rule may violate FPJR, so neither of EJR and FPJR implies the other.
2501.12016
Are Traditional Deep Learning Model Approaches as Effective as a Retinal-Specific Foundation Model for Ocular and Systemic Disease Detection?
cs.CV cs.LG
Background: RETFound, a self-supervised, retina-specific foundation model (FM), showed potential in downstream applications. However, its comparative performance with traditional deep learning (DL) models remains incompletely understood. This study aimed to evaluate RETFound against three ImageNet-pretrained supervised DL models (ResNet50, ViT-base, SwinV2) in detecting ocular and systemic diseases. Methods: We fine-tuned/trained RETFound and three DL models on full datasets, 50%, 20%, and fixed sample sizes (400, 200, 100 images, with half comprising disease cases; for each DR severity class, 100 and 50 cases were used. Fine-tuned models were tested internally using the SEED (53,090 images) and APTOS-2019 (3,672 images) datasets and externally validated on population-based (BES, CIEMS, SP2, UKBB) and open-source datasets (ODIR-5k, PAPILA, GAMMA, IDRiD, MESSIDOR-2). Model performance was compared using area under the receiver operating characteristic curve (AUC) and Z-tests with Bonferroni correction (P<0.05/3). Interpretation: Traditional DL models are mostly comparable to RETFound for ocular disease detection with large datasets. However, RETFound is superior in systemic disease detection with smaller datasets. These findings offer valuable insights into the respective merits and limitation of traditional models and FMs.
2501.12020
On the "Illusion" of Gender Bias in Face Recognition: Explaining the Fairness Issue Through Non-demographic Attributes
cs.CV
Face recognition systems (FRS) exhibit significant accuracy differences based on the user's gender. Since such a gender gap reduces the trustworthiness of FRS, more recent efforts have tried to find the causes. However, these studies make use of manually selected, correlated, and small-sized sets of facial features to support their claims. In this work, we analyse gender bias in face recognition by successfully extending the search domain to decorrelated combinations of 40 non-demographic facial characteristics. First, we propose a toolchain to effectively decorrelate and aggregate facial attributes to enable a less-biased gender analysis on large-scale data. Second, we introduce two new fairness metrics to measure fairness with and without context. Based on these grounds, we thirdly present a novel unsupervised algorithm able to reliably identify attribute combinations that lead to vanishing bias when used as filter predicates for balanced testing datasets. The experiments show that the gender gap vanishes when images of male and female subjects share specific attributes, clearly indicating that the issue is not a question of biology but of the social definition of appearance. These findings could reshape our understanding of fairness in face biometrics and provide insights into FRS, helping to address gender bias issues.
2501.12022
Foreign object segmentation in chest x-rays through anatomy-guided shape insertion
cs.CV
In this paper, we tackle the challenge of instance segmentation for foreign objects in chest radiographs, commonly seen in postoperative follow-ups with stents, pacemakers, or ingested objects in children. The diversity of foreign objects complicates dense annotation, as shown in insufficient existing datasets. To address this, we propose the simple generation of synthetic data through (1) insertion of arbitrary shapes (lines, polygons, ellipses) with varying contrasts and opacities, and (2) cut-paste augmentations from a small set of semi-automatically extracted labels. These insertions are guided by anatomy labels to ensure realistic placements, such as stents appearing only in relevant vessels. Our approach enables networks to segment complex structures with minimal manually labeled data. Notably, it achieves performance comparable to fully supervised models while using 93\% fewer manual annotations.
2501.12023
Comparative Analysis of Pre-trained Deep Learning Models and DINOv2 for Cushing's Syndrome Diagnosis in Facial Analysis
cs.LG cs.CV eess.IV
Cushing's syndrome is a condition caused by excessive glucocorticoid secretion from the adrenal cortex, often manifesting with moon facies and plethora, making facial data crucial for diagnosis. Previous studies have used pre-trained convolutional neural networks (CNNs) for diagnosing Cushing's syndrome using frontal facial images. However, CNNs are better at capturing local features, while Cushing's syndrome often presents with global facial features. Transformer-based models like ViT and SWIN, which utilize self-attention mechanisms, can better capture long-range dependencies and global features. Recently, DINOv2, a foundation model based on visual Transformers, has gained interest. This study compares the performance of various pre-trained models, including CNNs, Transformer-based models, and DINOv2, in diagnosing Cushing's syndrome. We also analyze gender bias and the impact of freezing mechanisms on DINOv2. Our results show that Transformer-based models and DINOv2 outperformed CNNs, with ViT achieving the highest F1 score of 85.74%. Both the pre-trained model and DINOv2 had higher accuracy for female samples. DINOv2 also showed improved performance when freezing parameters. In conclusion, Transformer-based models and DINOv2 are effective for Cushing's syndrome classification.
2501.12025
Low-Cost 3D printed, Biocompatible Ionic Polymer Membranes for Soft Actuators
cond-mat.soft cs.RO
Ionic polymer actuators, in essence, consist of ion exchange polymers sandwiched between layers of electrodes. They have recently gained recognition as promising candidates for soft actuators due to their lightweight nature, noise-free operation, and low-driving voltages. However, the materials traditionally utilized to develop them are often not human/environmentally friendly. Thus, to address this issue, researchers have been focusing on developing biocompatible versions of this actuator. Despite this, such actuators still face challenges in achieving high performance, in payload capacity, bending capabilities, and response time. In this paper, we present a biocompatible ionic polymer actuator whose membrane is fully 3D printed utilizing a direct ink writing method. The structure of the printed membranes consists of biodegradable ionic fluid encapsulated within layers of activated carbon polymers. From the microscopic observations of its structure, we confirmed that the ionic polymer is well encapsulated. The actuators can achieve a bending performance of up to 124$^\circ$ (curvature of 0.82 $\text{cm}^{-1}$), which, to our knowledge, is the highest curvature attained by any bending ionic polymer actuator to date. It can operate comfortably up to a 2 Hz driving frequency and can achieve blocked forces of up to 0.76 mN. Our results showcase a promising, high-performing biocompatible ionic polymer actuator, whose membrane can be easily manufactured in a single step using a standard FDM 3D printer. This approach paves the way for creating customized designs for functional soft robotic applications, including human-interactive devices, in the near future.
2501.12030
Advancing Earth Observation: A Survey on AI-Powered Image Processing in Satellites
cs.LG cs.CV
Advancements in technology and reduction in it's cost have led to a substantial growth in the quality & quantity of imagery captured by Earth Observation (EO) satellites. This has presented a challenge to the efficacy of the traditional workflow of transmitting this imagery to Earth for processing. An approach to addressing this issue is to use pre-trained artificial intelligence models to process images on-board the satellite, but this is difficult given the constraints within a satellite's environment. This paper provides an up-to-date and thorough review of research related to image processing on-board Earth observation satellites. The significant constraints are detailed along with the latest strategies to mitigate them.
2501.12032
Multi-Tenant SmartNICs for In-Network Preprocessing of Recommender Systems
cs.AR cs.DC cs.LG
Keeping ML-based recommender models up-to-date as data drifts and evolves is essential to maintain accuracy. As a result, online data preprocessing plays an increasingly important role in serving recommender systems. Existing solutions employ multiple CPU workers to saturate the input bandwidth of a single training node. Such an approach results in high deployment costs and energy consumption. For instance, a recent report from industrial deployments shows that data storage and ingestion pipelines can account for over 60\% of the power consumption in a recommender system. In this paper, we tackle the issue from a hardware perspective by introducing Piper, a flexible and network-attached accelerator that executes data loading and preprocessing pipelines in a streaming fashion. As part of the design, we define MiniPipe, the smallest pipeline unit enabling multi-pipeline implementation by executing various data preprocessing tasks across the single board, giving Piper the ability to be reconfigured at runtime. Our results, using publicly released commercial pipelines, show that Piper, prototyped on a power-efficient FPGA, achieves a 39$\sim$105$\times$ speedup over a server-grade, 128-core CPU and 3$\sim$17$\times$ speedup over GPUs like RTX 3090 and A100 in multiple pipelines. The experimental analysis demonstrates that Piper provides advantages in both latency and energy efficiency for preprocessing tasks in recommender systems, providing an alternative design point for systems that today are in very high demand.
2501.12033
Harnessing Generative Pre-Trained Transformer for Datacenter Packet Trace Generation
cs.NI cs.AI
Today, the rapid growth of applications reliant on datacenters calls for new advancements to meet the increasing traffic and computational demands. Traffic traces from datacenters are essential for further development and optimization of future datacenters. However, traces are rarely released to the public. Researchers often use simplified mathematical models that lack the depth needed to recreate intricate traffic patterns and, thus, miss optimization opportunities found in realistic traffic. In this preliminary work, we introduce DTG-GPT, a packet-level Datacenter Traffic Generator (DTG), based on the generative pre-trained transformer (GPT) architecture used by many state-of-the-art large language models. We train our model on a small set of available traffic traces from different domains and offer a simple methodology to evaluate the fidelity of the generated traces to their original counterparts. We show that DTG-GPT can synthesize novel traces that mimic the spatiotemporal patterns found in real traffic traces. We further demonstrate that DTG-GPT can generate traces for networks of different scales while maintaining fidelity. Our findings indicate the potential that, in the future, similar models to DTG-GPT will allow datacenter operators to release traffic information to the research community via trained GPT models.
2501.12040
Select2Drive: Pragmatic Communications for Real-Time Collaborative Autonomous Driving
cs.CE
Vehicle-to-Everything communications-assisted Autonomous Driving (V2X-AD) has witnessed remarkable advancements in recent years, with pragmatic communications (PragComm) emerging as a promising paradigm for real-time collaboration among vehicles and other agents.Simultaneously, extensive research has explored the interplay between collaborative perception and decision-making in end-to-end driving frameworks.In this work, we revisit the collaborative driving problem and propose the Select2Drive framework to optimize the utilization of limited computational and communication resources.Particularly, to mitigate cumulative latency in perception and decision-making, Select2Drive introduces Distributed Predictive Perception (DPP) by formulating an active prediction paradigm and simplifies high-dimensional semantic feature prediction into computation cost-efficient, motion-aware reconstruction. Given the "less is more" principle that a broadened perceptual horizon possibly confuses the decision module rather than contributing to it, Select2Drive utilizes Area-of-Importance-based PragComm (APC) to prioritize the communications of critical regions, thus boosting both communication efficiency and decision-making efficacy. Empirical evaluations on the V2Xverse dataset and CARLA driving simulator demonstrate that Select2Drive achieves a 11.31% (resp. 7.69%) improvement in offline perception tasks under limited bandwidth (resp. pose error conditions). Moreover, it delivers at most 14.68% and 31.76% enhancement in closed-loop driving scores and route completion rates, particularly in scenarios characterized by dense traffic and high-speed dynamics.
2501.12043
High-Fidelity Coherent-One-Way QKD Simulation Framework for 6G Networks: Bridging Theory and Reality
quant-ph cs.SY eess.SY
Quantum key distribution (QKD) has been emerged as a promising solution for guaranteeing information-theoretic security. Inspired by this, a great amount of research effort has been recently put on designing and testing QKD systems as well as articulating preliminary application scenarios. However, due to the considerable high-cost of QKD equipment, a lack of QKD communication system design tools, wide deployment of such systems and networks is challenging. Motivated by this, this paper introduces a QKD communication system design tool. First we articulate key operation elements of the QKD, and explain the feasibility and applicability of coherent-one-way (COW) QKD solutions. Next, we focus on documenting the corresponding simulation framework as well as defining the key performance metrics, i.e., quantum bit error rate (QBER), and secrecy key rate. To verify the accuracy of the simulation framework, we design and deploy a real-world QKD setup. We perform extensive experiments for three deployments of diverse transmission distance in the presence or absence of a QKD eavesdropper. The results reveal an acceptable match between simulations and experiments rendering the simulation framework a suitable tool for QKD communication system design.
2501.12046
Communication-Efficient and Privacy-Adaptable Mechanism for Federated Learning
cs.LG
Training machine learning models on decentralized private data via federated learning (FL) poses two key challenges: communication efficiency and privacy protection. In this work, we address these challenges within the trusted aggregator model by introducing a novel approach called the Communication-Efficient and Privacy-Adaptable Mechanism (CEPAM), achieving both objectives simultaneously. In particular, CEPAM leverages the rejection-sampled universal quantizer (RSUQ), a construction of randomized vector quantizer whose resulting distortion is equivalent to a prescribed noise, such as Gaussian or Laplace noise, enabling joint differential privacy and compression. Moreover, we analyze the trade-offs among user privacy, global utility, and transmission rate of CEPAM by defining appropriate metrics for FL with differential privacy and compression. Our CEPAM provides the additional benefit of privacy adaptability, allowing clients and the server to customize privacy protection based on required accuracy and protection. We assess CEPAM's utility performance using MNIST dataset, demonstrating that CEPAM surpasses baseline models in terms of learning accuracy.
2501.12048
Adaptive Class Learning to Screen Diabetic Disorders in Fundus Images of Eye
cs.CV cs.AI
The prevalence of ocular illnesses is growing globally, presenting a substantial public health challenge. Early detection and timely intervention are crucial for averting visual impairment and enhancing patient prognosis. This research introduces a new framework called Class Extension with Limited Data (CELD) to train a classifier to categorize retinal fundus images. The classifier is initially trained to identify relevant features concerning Healthy and Diabetic Retinopathy (DR) classes and later fine-tuned to adapt to the task of classifying the input images into three classes: Healthy, DR, and Glaucoma. This strategy allows the model to gradually enhance its classification capabilities, which is beneficial in situations where there are only a limited number of labeled datasets available. Perturbation methods are also used to identify the input image characteristics responsible for influencing the models decision-making process. We achieve an overall accuracy of 91% on publicly available datasets.
2501.12050
Representation Learning with Parameterised Quantum Circuits for Advancing Speech Emotion Recognition
cs.LG cs.SD eess.AS
Speech Emotion Recognition (SER) is a complex and challenging task in human-computer interaction due to the intricate dependencies of features and the overlapping nature of emotional expressions conveyed through speech. Although traditional deep learning methods have shown effectiveness, they often struggle to capture subtle emotional variations and overlapping states. This paper introduces a hybrid classical-quantum framework that integrates Parameterised Quantum Circuits (PQCs) with conventional Convolutional Neural Network (CNN) architectures. By leveraging quantum properties such as superposition and entanglement, the proposed model enhances feature representation and captures complex dependencies more effectively than classical methods. Experimental evaluations conducted on benchmark datasets, including IEMOCAP, RECOLA, and MSP-Improv, demonstrate that the hybrid model achieves higher accuracy in both binary and multi-class emotion classification while significantly reducing the number of trainable parameters. While a few existing studies have explored the feasibility of using Quantum Circuits to reduce model complexity, none have successfully shown how they can enhance accuracy. This study is the first to demonstrate that Quantum Circuits has the potential to improve the accuracy of SER. The findings highlight the promise of QML to transform SER, suggesting a promising direction for future research and practical applications in emotion-aware systems.
2501.12051
MedS$^3$: Towards Medical Small Language Models with Self-Evolved Slow Thinking
cs.CL
Medical language models (MLMs) have become pivotal in advancing medical natural language processing. However, prior models that rely on pre-training or supervised fine-tuning often exhibit low data efficiency and limited practicality in real-world clinical applications. While OpenAI's o1 highlights test-time scaling in mathematics, attempts to replicate this approach in medicine typically distill responses from GPT-series models to open-source models, focusing primarily on multiple-choice tasks. This strategy, though straightforward, neglects critical concerns like data privacy and realistic deployment in clinical settings. In this work, we present a deployable, small-scale medical reasoning system, MedS3, designed for long-chain reasoning in clinical tasks using a self-evolution paradigm. Starting with a seed dataset of around 8,000 instances spanning five domains and 16 datasets, we prompt a base policy model to perform Monte Carlo Tree Search (MCTS) to construct rule-verifiable reasoning chains. Each reasoning step is assigned an evolution rollout value, allowing verified trajectories to train the policy model and the process reward model (PRM). During inference, the policy model generates multiple responses, and the reward model selects the one with a newly proposed PRM-guided Vote-Sum (P-VS) strategy. Experiments on eleven evaluation datasets demonstrate that MedS3 outperforms not only the prior strongest medical model by 6.59, but also 32B-level general reasoning models by 8.71 points. Code and data are available at https://github.com/pixas/MedSSS.
2501.12052
Aggrotech: Leveraging Deep Learning for Sustainable Tomato Disease Management
cs.CV cs.LG
Tomato crop health plays a critical role in ensuring agricultural productivity and food security. Timely and accurate detection of diseases affecting tomato plants is vital for effective disease management. In this study, we propose a deep learning-based approach for Tomato Leaf Disease Detection using two well-established convolutional neural networks (CNNs), namely VGG19 and Inception v3. The experiment is conducted on the Tomato Villages Dataset, encompassing images of both healthy tomato leaves and leaves afflicted by various diseases. The VGG19 model is augmented with fully connected layers, while the Inception v3 model is modified to incorporate a global average pooling layer and a dense classification layer. Both models are trained on the prepared dataset, and their performances are evaluated on a separate test set. This research employs VGG19 and Inception v3 models on the Tomato Villages dataset (4525 images) for tomato leaf disease detection. The models' accuracy of 93.93% with dropout layers demonstrates their usefulness for crop health monitoring. The paper suggests a deep learning-based strategy that includes normalization, resizing, dataset preparation, and unique model architectures. During training, VGG19 and Inception v3 serve as feature extractors, with possible data augmentation and fine-tuning. Metrics like accuracy, precision, recall, and F1 score are obtained through evaluation on a test set and offer important insights into the strengths and shortcomings of the model. The method has the potential for practical use in precision agriculture and could help tomato crops prevent illness early on.
2501.12053
PINNsAgent: Automated PDE Surrogation with Large Language Models
cs.CE
Solving partial differential equations (PDEs) using neural methods has been a long-standing scientific and engineering research pursuit. Physics-Informed Neural Networks (PINNs) have emerged as a promising alternative to traditional numerical methods for solving PDEs. However, the gap between domain-specific knowledge and deep learning expertise often limits the practical application of PINNs. Previous works typically involve manually conducting extensive PINNs experiments and summarizing heuristic rules for hyperparameter tuning. In this work, we introduce PINNsAgent, a novel surrogation framework that leverages large language models (LLMs) and utilizes PINNs as a foundation to bridge the gap between domain-specific knowledge and deep learning. Specifically, PINNsAgent integrates (1) Physics-Guided Knowledge Replay (PGKR), which encodes the essential characteristics of PDEs and their associated best-performing PINNs configurations into a structured format, enabling efficient knowledge transfer from solved PDEs to similar problems and (2) Memory Tree Reasoning, a strategy that effectively explores the search space for optimal PINNs architectures. By leveraging LLMs and exploration strategies, PINNsAgent enhances the automation and efficiency of PINNs-based solutions. We evaluate PINNsAgent on 14 benchmark PDEs, demonstrating its effectiveness in automating the surrogation process and significantly improving the accuracy of PINNs-based solutions.
2501.12054
ORCAst: Operational High-Resolution Current Forecasts
cs.CV physics.ao-ph
We present ORCAst, a multi-stage, multi-arm network for Operational high-Resolution Current forecAsts over one week. Producing real-time nowcasts and forecasts of ocean surface currents is a challenging problem due to indirect or incomplete information from satellite remote sensing data. Entirely trained on real satellite data and in situ measurements from drifters, our model learns to forecast global ocean surface currents using various sources of ground truth observations in a multi-stage learning procedure. Our multi-arm encoder-decoder model architecture allows us to first predict sea surface height and geostrophic currents from larger quantities of nadir and SWOT altimetry data, before learning to predict ocean surface currents from much more sparse in situ measurements from drifters. Training our model on specific regions improves performance. Our model achieves stronger nowcast and forecast performance in predicting ocean surface currents than various state-of-the-art methods.
2501.12057
Unified 3D MRI Representations via Sequence-Invariant Contrastive Learning
cs.CV physics.med-ph
Self-supervised deep learning has accelerated 2D natural image analysis but remains difficult to translate into 3D MRI, where data are scarce and pre-trained 2D backbones cannot capture volumetric context. We present a sequence-invariant self-supervised framework leveraging quantitative MRI (qMRI). By simulating multiple MRI contrasts from a single 3D qMRI scan and enforcing consistent representations across these contrasts, we learn anatomy-centric rather than sequence-specific features. This yields a robust 3D encoder that performs strongly across varied tasks and protocols. Experiments on healthy brain segmentation (IXI), stroke lesion segmentation (ARC), and MRI denoising show significant gains over baseline SSL approaches, especially in low-data settings (up to +8.3% Dice, +4.2 dB PSNR). Our model also generalises effectively to unseen sites, demonstrating potential for more scalable and clinically reliable volumetric analysis. All code and trained models are publicly available.
2501.12058
Fractional Subadditivity of Submodular Functions: Equality Conditions and Their Applications
cs.IT math.IT
Submodular functions are known to satisfy various forms of fractional subadditivity. This work investigates the conditions for equality to hold exactly or approximately in the fractional subadditivity of submodular functions. We establish that a small gap in the inequality implies that the function is close to being modular, and that the gap is zero if and only if the function is modular. We then present natural implications of these results for special cases of submodular functions, such as entropy, relative entropy, and matroid rank. As a consequence, we characterize the necessary and sufficient conditions for equality to hold in Shearer's lemma, recovering a result of Ellis \emph{et al.} (2016) as a special case. We leverage our results to propose a new multivariate mutual information, which generalizes Watanabe's total correlation (1960), Han's dual total correlation (1978), and Csisz\'ar and Narayan's shared information (2004), and analyze its properties. Among these properties, we extend Watanabe's characterization of total correlation as the maximum correlation over partitions to fractional partitions. When applied to matrix determinantal inequalities for positive definite matrices, our results recover the equality conditions of the classical determinantal inequalities of Hadamard, Sz\'asz, and Fischer as special cases.
2501.12060
GSVC: Efficient Video Representation and Compression Through 2D Gaussian Splatting
cs.CV cs.MM
3D Gaussian splats have emerged as a revolutionary, effective, learned representation for static 3D scenes. In this work, we explore using 2D Gaussian splats as a new primitive for representing videos. We propose GSVC, an approach to learning a set of 2D Gaussian splats that can effectively represent and compress video frames. GSVC incorporates the following techniques: (i) To exploit temporal redundancy among adjacent frames, which can speed up training and improve the compression efficiency, we predict the Gaussian splats of a frame based on its previous frame; (ii) To control the trade-offs between file size and quality, we remove Gaussian splats with low contribution to the video quality; (iii) To capture dynamics in videos, we randomly add Gaussian splats to fit content with large motion or newly-appeared objects; (iv) To handle significant changes in the scene, we detect key frames based on loss differences during the learning process. Experiment results show that GSVC achieves good rate-distortion trade-offs, comparable to state-of-the-art video codecs such as AV1 and VVC, and a rendering speed of 1500 fps for a 1920x1080 video.
2501.12061
Tackling Uncertainties in Multi-Agent Reinforcement Learning through Integration of Agent Termination Dynamics
cs.LG cs.MA
Multi-Agent Reinforcement Learning (MARL) has gained significant traction for solving complex real-world tasks, but the inherent stochasticity and uncertainty in these environments pose substantial challenges to efficient and robust policy learning. While Distributional Reinforcement Learning has been successfully applied in single-agent settings to address risk and uncertainty, its application in MARL is substantially limited. In this work, we propose a novel approach that integrates distributional learning with a safety-focused loss function to improve convergence in cooperative MARL tasks. Specifically, we introduce a Barrier Function based loss that leverages safety metrics, identified from inherent faults in the system, into the policy learning process. This additional loss term helps mitigate risks and encourages safer exploration during the early stages of training. We evaluate our method in the StarCraft II micromanagement benchmark, where our approach demonstrates improved convergence and outperforms state-of-the-art baselines in terms of both safety and task completion. Our results suggest that incorporating safety considerations can significantly enhance learning performance in complex, multi-agent environments.
2501.12066
The Generalized Chernoff-Stein Lemma, Applications and Examples
cs.IT math.IT
In this manuscript we define the notion of "$\delta$-typicality" for both entropy and relative entropy, as well as a notion of $\epsilon$-goodness and provide an extension to Stein's lemma for continuous quantities as well as correlated setups. We apply the derived results on the Gaussian hypothesis testing problem where the observations are possibly correlated.
2501.12067
EDoRA: Efficient Weight-Decomposed Low-Rank Adaptation via Singular Value Decomposition
cs.LG cs.AI cs.CL
Parameter-efficient fine-tuning methods, such as LoRA, reduces the number of trainable parameters. However, they often suffer from scalability issues and differences between their learning pattern and full fine-tuning. To overcome these limitations, we propose Efficient Weight-Decomposed Low-Rank Adaptation (EDoRA): a novel PEFT method that decomposes pre-trained weights into magnitude and directional components. By freezing low-rank matrices, initializing them by singular value decomposition, and introducing a small trainable matrix between them, EDoRA achieves substantial reduction in trainable parameters while maintaining learning capacity. Experimental results on the GLUE benchmark demonstrate that EDoRA achieves competitive or superior performance compared to state-of-the-art methods, such as LoRA and DoRA, with up to 30x fewer trainable parameters. This makes EDoRA a highly efficient solution for adapting LLMs to diverse tasks under memory-constrained settings. Code is available at https://github.com/Hamid-Nasiri/EDoRA .
2501.12071
Co-Paced Learning Strategy Based on Confidence for Flying Bird Object Detection Model Training
cs.CV
To mitigate the adverse effects of hard samples on the training of the Flying Bird Object Detection (FBOD) model for surveillance videos, we propose a Co-Paced Learning Based on Confidence (CPL-BC) strategy and apply this strategy to the training process of the FBOD model. This strategy involves maintaining two models with identical structures but different initial parameter configurations, which collaborate with each other to select easy samples with prediction confidence exceeding a set threshold for training. As training progresses, the strategy gradually lowers the threshold, allowing more samples to participate, enhancing the model's ability to recognize objects from easy to hard. Before applying the CPL-BC strategy to train the FBOD models, we initially trained the two FBOD models to equip them with the capability to assess the difficulty level of flying bird object samples. Experimental results on two different datasets of flying bird objects in surveillance videos demonstrate that, compared to other model learning strategies, CPL-BC significantly improves detection accuracy, verifying the effectiveness and advancement of this method.
2501.12072
Fault-tolerance of [[6, 1, 3]] non-CSS code family generated using measurements on graph states
quant-ph cs.IT math.IT
We construct and analyze the fault tolerance of $[[6,1,3]]$ non-CSS quantum error correcting code under the anisotropic and depolarizing noise models. This rate-optimized code achieves fault-tolerance using a single ancilla qubit for syndrome measurement under anisotropic noise conditions. This method was called fault-tolerance using bare ancilla by Brown \emph{et al.} We give explicit construction of the code using measurements on non-planar graph states. We also argue that using our approach, we can construct a family of such fault-tolerant codes. This method fills a notable gap in constructing fault-tolerant non-CSS code families.
2501.12073
Towards autonomous photogrammetric forest inventory using a lightweight under-canopy robotic drone
cs.RO cs.CV
Drones are increasingly used in forestry to capture high-resolution remote sensing data. While operations above the forest canopy are already highly automated, flying inside forests remains challenging, primarily relying on manual piloting. Inside dense forests, reliance on the Global Navigation Satellite System (GNSS) for localization is not feasible. Additionally, the drone must autonomously adjust its flight path to avoid collisions. Recently, advancements in robotics have enabled autonomous drone flights in GNSS-denied obstacle-rich areas. In this article, a step towards autonomous forest data collection is taken by building a prototype of a robotic under-canopy drone utilizing state-of-the-art open-source methods and validating its performance for data collection inside forests. The autonomous flight capability was evaluated through multiple test flights in two boreal forest test sites. The tree parameter estimation capability was studied by conducting diameter at breast height (DBH) estimation using onboard stereo camera data and photogrammetric methods. The prototype conducted flights in selected challenging forest environments, and the experiments showed excellent performance in forest reconstruction with a miniaturized stereoscopic photogrammetric system. The stem detection algorithm managed to identify 79.31 % of the stems. The DBH estimation had a root mean square error (RMSE) of 3.33 cm (12.79 %) and a bias of 1.01 cm (3.87 %) across all trees. For trees with a DBH less than 30 cm, the RMSE was 1.16 cm (5.74 %), and the bias was 0.13 cm (0.64 %). When considering the overall performance in terms of DBH accuracy, autonomy, and forest complexity, the proposed approach was superior compared to methods proposed in the scientific literature. Results provided valuable insights into autonomous forest reconstruction using drones, and several further development topics were proposed.
2501.12074
Optimizing Portfolio Performance through Clustering and Sharpe Ratio-Based Optimization: A Comparative Backtesting Approach
cs.LG q-fin.PM
Optimizing portfolio performance is a fundamental challenge in financial modeling, requiring the integration of advanced clustering techniques and data-driven optimization strategies. This paper introduces a comparative backtesting approach that combines clustering-based portfolio segmentation and Sharpe ratio-based optimization to enhance investment decision-making. First, we segment a diverse set of financial assets into clusters based on their historical log-returns using K-Means clustering. This segmentation enables the grouping of assets with similar return characteristics, facilitating targeted portfolio construction. Next, for each cluster, we apply a Sharpe ratio-based optimization model to derive optimal weights that maximize risk-adjusted returns. Unlike traditional mean-variance optimization, this approach directly incorporates the trade-off between returns and volatility, resulting in a more balanced allocation of resources within each cluster. The proposed framework is evaluated through a backtesting study using historical data spanning multiple asset classes. Optimized portfolios for each cluster are constructed and their cumulative returns are compared over time against a traditional equal-weighted benchmark portfolio.
2501.12076
From Niche to Mainstream: Community Size and Engagement in Social Media Conversations
cs.SI cs.CY
The architecture of public discourse has been profoundly reshaped by social media platforms, which mediate interactions at an unprecedented scale and complexity. This study analyzes user behavior across six platforms over 33 years, exploring how the size of conversations and communities influences dialogue dynamics. Our findings reveal that smaller platforms foster richer, more sustained interactions, while larger platforms drive broader but shorter participation. Moreover, we observe that the propensity for users to re-engage in a conversation decreases as community size grows, with niche environments as a notable exception, where participation remains robust. These findings show an interdependence between platform architecture, user engagement, and community dynamics, shedding light on how digital ecosystems shape the structure and quality of public discourse.
2501.12082
A Multi-annotated and Multi-modal Dataset for Wide-angle Video Quality Assessment
cs.CV eess.IV
Wide-angle video is favored for its wide viewing angle and ability to capture a large area of scenery, making it an ideal choice for sports and adventure recording. However, wide-angle video is prone to deformation, exposure and other distortions, resulting in poor video quality and affecting the perception and experience, which may seriously hinder its application in fields such as competitive sports. Up to now, few explorations focus on the quality assessment issue of wide-angle video. This deficiency primarily stems from the absence of a specialized dataset for wide-angle videos. To bridge this gap, we construct the first Multi-annotated and multi-modal Wide-angle Video quality assessment (MWV) dataset. Then, the performances of state-of-the-art video quality methods on the MWV dataset are investigated by inter-dataset testing and intra-dataset testing. Experimental results show that these methods impose significant limitations on their applicability.
2501.12085
Scalable Whole Slide Image Representation Using K-Mean Clustering and Fisher Vector Aggregation
cs.CV cs.AI cs.LG
Whole slide images (WSIs) are high-resolution, gigapixel sized images that pose significant computational challenges for traditional machine learning models due to their size and heterogeneity.In this paper, we present a scalable and efficient methodology for WSI classification by leveraging patch-based feature extraction, clustering, and Fisher vector encoding. Initially, WSIs are divided into fixed size patches, and deep feature embeddings are extracted from each patch using a pre-trained convolutional neural network (CNN). These patch-level embeddings are subsequently clustered using K-means clustering, where each cluster aggregates semantically similar regions of the WSI. To effectively summarize each cluster, Fisher vector representations are computed by modeling the distribution of patch embeddings in each cluster as a parametric Gaussian mixture model (GMM). The Fisher vectors from each cluster are concatenated into a high-dimensional feature vector, creating a compact and informative representation of the entire WSI. This feature vector is then used by a classifier to predict the WSI's diagnostic label. Our method captures local and global tissue structures and yields robust performance for large-scale WSI classification, demonstrating superior accuracy and scalability compared to other approaches.
2501.12086
DSTSA-GCN: Advancing Skeleton-Based Gesture Recognition with Semantic-Aware Spatio-Temporal Topology Modeling
cs.CV
Graph convolutional networks (GCNs) have emerged as a powerful tool for skeleton-based action and gesture recognition, thanks to their ability to model spatial and temporal dependencies in skeleton data. However, existing GCN-based methods face critical limitations: (1) they lack effective spatio-temporal topology modeling that captures dynamic variations in skeletal motion, and (2) they struggle to model multiscale structural relationships beyond local joint connectivity. To address these issues, we propose a novel framework called Dynamic Spatial-Temporal Semantic Awareness Graph Convolutional Network (DSTSA-GCN). DSTSA-GCN introduces three key modules: Group Channel-wise Graph Convolution (GC-GC), Group Temporal-wise Graph Convolution (GT-GC), and Multi-Scale Temporal Convolution (MS-TCN). GC-GC and GT-GC operate in parallel to independently model channel-specific and frame-specific correlations, enabling robust topology learning that accounts for temporal variations. Additionally, both modules employ a grouping strategy to adaptively capture multiscale structural relationships. Complementing this, MS-TCN enhances temporal modeling through group-wise temporal convolutions with diverse receptive fields. Extensive experiments demonstrate that DSTSA-GCN significantly improves the topology modeling capabilities of GCNs, achieving state-of-the-art performance on benchmark datasets for gesture and action recognition, including SHREC17 Track, DHG-14\/28, NTU-RGB+D, and NTU-RGB+D-120.
2501.12087
UAV-Assisted Real-Time Disaster Detection Using Optimized Transformer Model
cs.CV
Disaster recovery and management present significant challenges, particularly in unstable environments and hard-to-reach terrains. These difficulties can be overcome by employing unmanned aerial vehicles (UAVs) equipped with onboard embedded platforms and camera sensors. In this work, we address the critical need for accurate and timely disaster detection by enabling onboard aerial imagery processing and avoiding connectivity, privacy, and latency issues despite the challenges posed by limited onboard hardware resources. We propose a UAV-assisted edge framework for real-time disaster management, leveraging our proposed model optimized for real-time aerial image classification. The optimization of the model employs post-training quantization techniques. For real-world disaster scenarios, we introduce a novel dataset, DisasterEye, featuring UAV-captured disaster scenes as well as ground-level images taken by individuals on-site. Experimental results demonstrate the effectiveness of our model, achieving high accuracy with reduced inference latency and memory usage on resource-constrained devices. The framework's scalability and adaptability make it a robust solution for real-time disaster detection on resource-limited UAV platforms.
2501.12092
Data-Aided Regularization of Direct-Estimate Combiner in Distributed MIMO Systems
eess.SP cs.IT math.IT
This paper explores the data-aided regularization of the direct-estimate combiner in the uplink of a distributed multiple-input multiple-output system. The network-wide combiner can be computed directly from the pilot signal received at each access point, eliminating the need for explicit channel estimation. However, the sample covariance matrix of the received pilot signal that is used in its computation may significantly deviate from the actual covariance matrix when the number of pilot symbols is limited. To address this, we apply a regularization to the sample covariance matrix using a shrinkage coefficient based on the received data signal. Initially, the shrinkage coefficient is determined by minimizing the difference between the sample covariance matrices obtained from the received pilot and data signals. Given the limitations of this approach in interference-limited scenarios, the shrinkage coefficient is iteratively optimized using the sample mean squared error of the hard-decision symbols, which is more closely related to the actual system's performance, e.g., the symbol error rate (SER). Numerical results demonstrate that the proposed regularization of the direct-estimate combiner significantly enhances the SER, particularly when the number of pilot symbols is limited.
2501.12102
Proxies for Distortion and Consistency with Applications for Real-World Image Restoration
cs.CV cs.AI cs.LG eess.IV
Real-world image restoration deals with the recovery of images suffering from an unknown degradation. This task is typically addressed while being given only degraded images, without their corresponding ground-truth versions. In this hard setting, designing and evaluating restoration algorithms becomes highly challenging. This paper offers a suite of tools that can serve both the design and assessment of real-world image restoration algorithms. Our work starts by proposing a trained model that predicts the chain of degradations a given real-world measured input has gone through. We show how this estimator can be used to approximate the consistency -- the match between the measurements and any proposed recovered image. We also use this estimator as a guiding force for the design of a simple and highly-effective plug-and-play real-world image restoration algorithm, leveraging a pre-trained diffusion-based image prior. Furthermore, this work proposes no-reference proxy measures of MSE and LPIPS, which, without access to the ground-truth images, allow ranking of real-world image restoration algorithms according to their (approximate) MSE and LPIPS. The proposed suite provides a versatile, first of its kind framework for evaluating and comparing blind image restoration algorithms in real-world scenarios.
2501.12104
Teacher Encoder-Student Decoder Denoising Guided Segmentation Network for Anomaly Detection
cs.CV cs.AI
Visual anomaly detection is a highly challenging task, often categorized as a one-class classification and segmentation problem. Recent studies have demonstrated that the student-teacher (S-T) framework effectively addresses this challenge. However, most S-T frameworks rely solely on pre-trained teacher networks to guide student networks in learning multi-scale similar features, overlooking the potential of the student networks to enhance learning through multi-scale feature fusion. In this study, we propose a novel model named PFADSeg, which integrates a pre-trained teacher network, a denoising student network with multi-scale feature fusion, and a guided anomaly segmentation network into a unified framework. By adopting a unique teacher-encoder and student-decoder denoising mode, the model improves the student network's ability to learn from teacher network features. Furthermore, an adaptive feature fusion mechanism is introduced to train a self-supervised segmentation network that synthesizes anomaly masks autonomously, significantly increasing detection performance. Evaluated on the MVTec AD dataset, PFADSeg achieves state-of-the-art results with an image-level AUC of 98.9%, a pixel-level mean precision of 76.4%, and an instance-level mean precision of 78.7%.
2501.12106
Can open source large language models be used for tumor documentation in Germany? -- An evaluation on urological doctors' notes
cs.CL cs.AI
Tumor documentation in Germany is largely done manually, requiring reading patient records and entering data into structured databases. Large language models (LLMs) could potentially enhance this process by improving efficiency and reliability. This evaluation tests eleven different open source LLMs with sizes ranging from 1-70 billion model parameters on three basic tasks of the tumor documentation process: identifying tumor diagnoses, assigning ICD-10 codes, and extracting the date of first diagnosis. For evaluating the LLMs on these tasks, a dataset of annotated text snippets based on anonymized doctors' notes from urology was prepared. Different prompting strategies were used to investigate the effect of the number of examples in few-shot prompting and to explore the capabilities of the LLMs in general. The models Llama 3.1 8B, Mistral 7B, and Mistral NeMo 12 B performed comparably well in the tasks. Models with less extensive training data or having fewer than 7 billion parameters showed notably lower performance, while larger models did not display performance gains. Examples from a different medical domain than urology could also improve the outcome in few-shot prompting, which demonstrates the ability of LLMs to handle tasks needed for tumor documentation. Open source LLMs show a strong potential for automating tumor documentation. Models from 7-12 billion parameters could offer an optimal balance between performance and resource efficiency. With tailored fine-tuning and well-designed prompting, these models might become important tools for clinical documentation in the future. The code for the evaluation is available from https://github.com/stefan-m-lenz/UroLlmEval. We also release the dataset as a new valuable resource that addresses the shortage of authentic and easily accessible benchmarks in German-language medical NLP.
2501.12113
Dual NUP Representations and Min-Maximization in Factor Graphs
stat.ML cs.LG cs.SY eess.SP eess.SY
Normals with unknown parameters (NUP) can be used to convert nontrivial model-based estimation problems into iterations of linear least-squares or Gaussian estimation problems. In this paper, we extend this approach by augmenting factor graphs with convex-dual variables and pertinent NUP representations. In particular, in a state space setting, we propose a new iterative forward-backward algorithm that is dual to a recently proposed backward-forward algorithm.
2501.12115
Meta-Sparsity: Learning Optimal Sparse Structures in Multi-task Networks through Meta-learning
cs.LG cs.CV
This paper presents meta-sparsity, a framework for learning model sparsity, basically learning the parameter that controls the degree of sparsity, that allows deep neural networks (DNNs) to inherently generate optimal sparse shared structures in multi-task learning (MTL) setting. This proposed approach enables the dynamic learning of sparsity patterns across a variety of tasks, unlike traditional sparsity methods that rely heavily on manual hyperparameter tuning. Inspired by Model Agnostic Meta-Learning (MAML), the emphasis is on learning shared and optimally sparse parameters in multi-task scenarios by implementing a penalty-based, channel-wise structured sparsity during the meta-training phase. This method improves the model's efficacy by removing unnecessary parameters and enhances its ability to handle both seen and previously unseen tasks. The effectiveness of meta-sparsity is rigorously evaluated by extensive experiments on two datasets, NYU-v2 and CelebAMask-HQ, covering a broad spectrum of tasks ranging from pixel-level to image-level predictions. The results show that the proposed approach performs well across many tasks, indicating its potential as a versatile tool for creating efficient and adaptable sparse neural networks. This work, therefore, presents an approach towards learning sparsity, contributing to the efforts in the field of sparse neural networks and suggesting new directions for research towards parsimonious models.
2501.12116
Efficient PINNs: Multi-Head Unimodular Regularization of the Solutions Space
cs.LG cs.AI hep-th math.AP
We present a machine learning framework to facilitate the solution of nonlinear multiscale differential equations and, especially, inverse problems using Physics-Informed Neural Networks (PINNs). This framework is based on what is called multihead (MH) training, which involves training the network to learn a general space of all solutions for a given set of equations with certain variability, rather than learning a specific solution of the system. This setup is used with a second novel technique that we call Unimodular Regularization (UR) of the latent space of solutions. We show that the multihead approach, combined with the regularization, significantly improves the efficiency of PINNs by facilitating the transfer learning process thereby enabling the finding of solutions for nonlinear, coupled, and multiscale differential equations.
2501.12118
Regularized dynamical parametric approximation of stiff evolution problems
math.NA cs.LG cs.NA
Evolutionary deep neural networks have emerged as a rapidly growing field of research. This paper studies numerical integrators for such and other classes of nonlinear parametrizations $ u(t) = \Phi(\theta(t)) $, where the evolving parameters $\theta(t)$ are to be computed. The primary focus is on tackling the challenges posed by the combination of stiff evolution problems and irregular parametrizations, which typically arise with neural networks, tensor networks, flocks of evolving Gaussians, and in further cases of overparametrization. We propose and analyse regularized parametric versions of the implicit Euler method and higher-order implicit Runge--Kutta methods for the time integration of the parameters in nonlinear approximations to evolutionary partial differential equations and large systems of stiff ordinary differential equations. At each time step, an ill-conditioned nonlinear optimization problem is solved approximately with a few regularized Gauss--Newton iterations. Error bounds for the resulting parametric integrator are derived by relating the computationally accessible Gauss--Newton iteration for the parameters to the computationally inaccessible Newton iteration for the underlying non-parametric time integration scheme. The theoretical findings are supported by numerical experiments that are designed to show key properties of the proposed parametric integrators.
2501.12119
ENTIRE: Learning-based Volume Rendering Time Prediction
cs.GR cs.CV cs.LG
We present ENTIRE, a novel approach for volume rendering time prediction. Time-dependent volume data from simulations or experiments typically comprise complex deforming structures across hundreds or thousands of time steps, which in addition to the camera configuration has a significant impact on rendering performance. We first extract a feature vector from a volume that captures its structure that is relevant for rendering time performance. Then we combine this feature vector with further relevant parameters (e.g. camera setup), and with this perform the final prediction. Our experiments conducted on various datasets demonstrate that our model is capable of efficiently achieving high prediction accuracy with fast response rates. We showcase ENTIRE's capability of enabling dynamic parameter adaptation for stable frame rates and load balancing in two case studies.
2501.12121
Learning Dynamic Representations via An Optimally-Weighted Maximum Mean Discrepancy Optimization Framework for Continual Learning
cs.LG cs.AI
Continual learning has emerged as a pivotal area of research, primarily due to its advantageous characteristic that allows models to persistently acquire and retain information. However, catastrophic forgetting can severely impair model performance. In this study, we address network forgetting by introducing a novel framework termed Optimally-Weighted Maximum Mean Discrepancy (OWMMD), which imposes penalties on representation alterations via a Multi-Level Feature Matching Mechanism (MLFMM). Furthermore, we propose an Adaptive Regularization Optimization (ARO) strategy to refine the adaptive weight vectors, which autonomously assess the significance of each feature layer throughout the optimization process, The proposed ARO approach can relieve the over-regularization problem and promote the future task learning. We conduct a comprehensive series of experiments, benchmarking our proposed method against several established baselines. The empirical findings indicate that our approach achieves state-of-the-art performance.
2501.12123
FedCLEAN: byzantine defense by CLustering Errors of Activation maps in Non-IID federated learning environments
cs.CR cs.AI
Federated Learning (FL) enables clients to collaboratively train a global model using their local datasets while reinforcing data privacy. However, FL is susceptible to poisoning attacks. Existing defense mechanisms assume that clients' data are independent and identically distributed (IID), making them ineffective in real-world applications where data are non-IID. This paper presents FedCLEAN, the first defense capable of filtering attackers' model updates in a non-IID FL environment. The originality of FedCLEAN is twofold. First, it relies on a client confidence score derived from the reconstruction errors of each client's model activation maps for a given trigger set, with reconstruction errors obtained by means of a Conditional Variational Autoencoder trained according to a novel server-side strategy. Second, we propose an ad-hoc trust propagation algorithm based on client scores, which allows building a cluster of benign clients while flagging potential attackers. Experimental results on the datasets MNIST and FashionMNIST demonstrate the robustness of FedCLEAN against Byzantine attackers in non-IID scenarios and a close-to-zero benign client misclassification rate, even in the absence of an attack.
2501.12124
On de Bruijn Array Codes Part II: Linear Codes
cs.IT math.IT
An M-sequence generated by a primitive polynomial has many interesting and desirable properties. A pseudo-random array is the two-dimensional generalization of an M-sequence. Similarly to primitive polynomials, there are irreducible and reducible polynomials whose all nonzero sequences have the same length. In this paper, a two-dimensional generalization for such sequences is given. This generalization is for a pseudo-random array code which is a set of $r_1 \times r_2$ arrays in which each $n_1 \times n_2$ nonzero matrix is contained exactly once as a window in one of the arrays. Moreover, these arrays have the shift-and-add property, i.e., the bitwise addition of two arrays (or a nontrivial shift of such arrays) is another array (or a shift of another array) from the code. All the known arrays can be formed by folding sequences generated from an irreducible polynomial or a reducible polynomial whose factors have the same degree and the same exponent. Two proof techniques are used to prove the parameters of the constructed arrays. The first one is based on another method for constructing some of these arrays. The second one is a generalization of a known proof technique. This generalization enables to present pseudo-random arrays with parameters not known before and also a variety of pseudo-random array codes which cannot be generated by the first method. The two techniques also suggest two different hierarchies between pseudo-random array codes. Finally, a method to verify whether a folding of sequences, generated by these polynomials, yields a pseudo-random array or a pseudo-random array code, will be presented.
2501.12125
Heterogeneous Federated Learning System for Sparse Healthcare Time-Series Prediction
cs.LG
In this paper, we propose a heterogeneous federated learning (HFL) system for sparse time series prediction in healthcare, which is a decentralized federated learning algorithm with heterogeneous transfers. We design dense and sparse feature tensors to deal with the sparsity of data sources. Heterogeneous federated learning is developed to share asynchronous parts of networks and select appropriate models for knowledge transfer. Experimental results show that the proposed HFL achieves the lowest prediction error among all benchmark systems on eight out of ten prediction tasks, with MSE reduction of 94.8%, 48.3%, and 52.1% compared to the benchmark systems. These results demonstrate the effectiveness of HFL in transferring knowledge from heterogeneous domains, especially in the smaller target domain. Ablation studies then demonstrate the effectiveness of the designed mechanisms for heterogeneous domain selection and switching in predicting healthcare time series with privacy, model security, and heterogeneous knowledge transfer.
2501.12128
Evaluating Efficiency and Engagement in Scripted and LLM-Enhanced Human-Robot Interactions
cs.RO cs.HC
To achieve natural and intuitive interaction with people, HRI frameworks combine a wide array of methods for human perception, intention communication, human-aware navigation and collaborative action. In practice, when encountering unpredictable behavior of people or unexpected states of the environment, these frameworks may lack the ability to dynamically recognize such states, adapt and recover to resume the interaction. Large Language Models (LLMs), owing to their advanced reasoning capabilities and context retention, present a promising solution for enhancing robot adaptability. This potential, however, may not directly translate to improved interaction metrics. This paper considers a representative interaction with an industrial robot involving approach, instruction, and object manipulation, implemented in two conditions: (1) fully scripted and (2) including LLM-enhanced responses. We use gaze tracking and questionnaires to measure the participants' task efficiency, engagement, and robot perception. The results indicate higher subjective ratings for the LLM condition, but objective metrics show that the scripted condition performs comparably, particularly in efficiency and focus during simple tasks. We also note that the scripted condition may have an edge over LLM-enhanced responses in terms of response latency and energy consumption, especially for trivial and repetitive interactions.
2501.12133
Distributed Multi-Head Learning Systems for Power Consumption Prediction
cs.LG
As more and more automatic vehicles, power consumption prediction becomes a vital issue for task scheduling and energy management. Most research focuses on automatic vehicles in transportation, but few focus on automatic ground vehicles (AGVs) in smart factories, which face complex environments and generate large amounts of data. There is an inevitable trade-off between feature diversity and interference. In this paper, we propose Distributed Multi-Head learning (DMH) systems for power consumption prediction in smart factories. Multi-head learning mechanisms are proposed in DMH to reduce noise interference and improve accuracy. Additionally, DMH systems are designed as distributed and split learning, reducing the client-to-server transmission cost, sharing knowledge without sharing local data and models, and enhancing the privacy and security levels. Experimental results show that the proposed DMH systems rank in the top-2 on most datasets and scenarios. DMH-E system reduces the error of the state-of-the-art systems by 14.5% to 24.0%. Effectiveness studies demonstrate the effectiveness of Pearson correlation-based feature engineering, and feature grouping with the proposed multi-head learning further enhances prediction performance.
2501.12135
Revisit the AWGN-goodness of Polar-like Lattices
cs.IT math.IT
This paper aims to provide a comprehensive introduction to lattices constructed based on polar-like codes and demonstrate some of their key properties, such as AWGN goodness. We first present polar lattices directly from the perspective of their generator matrix. Next, we discuss their connection with the recently proposed PAC (polarization adjusted convolutional) lattices and analyze the structural advantages of PAC lattices, through which the AWGN-goodness of PAC lattices can be conveniently demonstrated.
2501.12136
Heterogeneous Federated Learning Systems for Time-Series Power Consumption Prediction with Multi-Head Embedding Mechanism
cs.LG
Time-series prediction is increasingly popular in a variety of applications, such as smart factories and smart transportation. Researchers have used various techniques to predict power consumption, but existing models lack discussion of collaborative learning and privacy issues among multiple clients. To address these issues, we propose Multi-Head Heterogeneous Federated Learning (MHHFL) systems that consist of multiple head networks, which independently act as carriers for federated learning. In the federated period, each head network is embedded into 2-dimensional vectors and shared with the centralized source pool. MHHFL then selects appropriate source networks and blends the head networks as knowledge transfer in federated learning. The experimental results show that the proposed MHHFL systems significantly outperform the benchmark and state-of-the-art systems and reduce the prediction error by 24.9% to 94.1%. The ablation studies demonstrate the effectiveness of the proposed mechanisms in the MHHFL (head network embedding and selection mechanisms), which significantly outperforms traditional federated average and random transfer.
2501.12147
Improving Influence-based Instruction Tuning Data Selection for Balanced Learning of Diverse Capabilities
cs.CL cs.AI cs.LG
Selecting appropriate training data is crucial for effective instruction fine-tuning of large language models (LLMs), which aims to (1) elicit strong capabilities, and (2) achieve balanced performance across a diverse range of tasks. Influence-based methods show promise in achieving (1) by estimating the contribution of each training example to the model's predictions, but often struggle with (2). Our systematic investigation reveals that this underperformance can be attributed to an inherent bias where certain tasks intrinsically have greater influence than others. As a result, data selection is often biased towards these tasks, not only hurting the model's performance on others but also, counterintuitively, harms performance on these high-influence tasks themselves. As a remedy, we propose BIDS, a Balanced and Influential Data Selection algorithm. BIDS first normalizes influence scores of the training data, and then iteratively balances data selection by choosing the training example with the highest influence on the most underrepresented task. Experiments with both Llama-3 and Mistral-v0.3 on seven benchmarks spanning five diverse capabilities show that BIDS consistently outperforms both state-of-the-art influence-based algorithms and other non-influence-based selection frameworks. Surprisingly, training on a 15% subset selected by BIDS can even outperform full-dataset training with a much more balanced performance. Our analysis further highlights the importance of both instance-level normalization and iterative optimization of selected data for balanced learning of diverse capabilities.
2501.12148
Deep Unfolding of Fixed-Point Based Algorithm for Weighted Sum Rate Maximization
cs.IT math.IT
In this paper, we propose a novel approach that harnesses the standard interference function, specifically tailored to address the unique challenges of non-convex optimization in wireless networks. We begin by establishing theoretical guarantees for our method under the assumption that the interference function exhibits log-concavity. Building on this foundation, we develop a Primal-Dual Algorithm (PDA) to approximate the solution to the Weighted Sum Rate (WSR) maximization problem. To further enhance computational efficiency, we leverage the deep unfolding technique, significantly reducing the complexity of the proposed algorithm. Through numerical experiments, we demonstrate the competitiveness of our method compared to the state-of-the-art fractional programming benchmark, commonly referred to as FPLinQ.
2501.12149
On the practical applicability of modern DFT functionals for chemical computations. Case study of DM21 applicability for geometry optimization
physics.comp-ph cond-mat.mtrl-sci cs.AI
Density functional theory (DFT) is probably the most promising approach for quantum chemistry calculations considering its good balance between calculations precision and speed. In recent years, several neural network-based functionals have been developed for exchange-correlation energy approximation in DFT, DM21 developed by Google Deepmind being the most notable between them. This study focuses on evaluating the efficiency of DM21 functional in predicting molecular geometries, with a focus on the influence of oscillatory behavior in neural network exchange-correlation functionals. We implemented geometry optimization in PySCF for the DM21 functional in geometry optimization problem, compared its performance with traditional functionals, and tested it on various benchmarks. Our findings reveal both the potential and the current challenges of using neural network functionals for geometry optimization in DFT. We propose a solution extending the practical applicability of such functionals and allowing to model new substances with their help.
2501.12150
DNRSelect: Active Best View Selection for Deferred Neural Rendering
cs.CV
Deferred neural rendering (DNR) is an emerging computer graphics pipeline designed for high-fidelity rendering and robotic perception. However, DNR heavily relies on datasets composed of numerous ray-traced images and demands substantial computational resources. It remains under-explored how to reduce the reliance on high-quality ray-traced images while maintaining the rendering fidelity. In this paper, we propose DNRSelect, which integrates a reinforcement learning-based view selector and a 3D texture aggregator for deferred neural rendering. We first propose a novel view selector for deferred neural rendering based on reinforcement learning, which is trained on easily obtained rasterized images to identify the optimal views. By acquiring only a few ray-traced images for these selected views, the selector enables DNR to achieve high-quality rendering. To further enhance spatial awareness and geometric consistency in DNR, we introduce a 3D texture aggregator that fuses pyramid features from depth maps and normal maps with UV maps. Given that acquiring ray-traced images is more time-consuming than generating rasterized images, DNRSelect minimizes the need for ray-traced data by using only a few selected views while still achieving high-fidelity rendering results. We conduct detailed experiments and ablation studies on the NeRF-Synthetic dataset to demonstrate the effectiveness of DNRSelect. The code will be released.
2501.12156
Characterization of Invariance, Periodic Solutions and Optimization of Dynamic Financial Networks
eess.SY cs.SY math.DS math.OC
Cascading failures, such as bankruptcies and defaults, pose a serious threat for the resilience of the global financial system. Indeed, because of the complex investment and cross-holding relations within the system, failures can occur as a result of the propagation of a financial collapse from one organization to another. While this problem has been studied in depth from a static angle, namely, when the system is at an equilibrium, we take a different perspective and study the corresponding dynamical system. The contribution of this paper is threefold. First, we carry out a systematic analysis of the regions of attraction and invariance of the system orthants, defined by the positive and negative values of the organizations' equity. Second, we investigate periodic solutions and show through a counterexample that there could exist periodic solutions of period greater than 2. Finally, we study the problem of finding the smallest cash injection that would bring the system to the maximal invariant region of the positive orthant.
2501.12157
Fast-RF-Shimming: Accelerate RF Shimming in 7T MRI using Deep Learning
cs.CV
Ultrahigh field (UHF) Magnetic Resonance Imaging (MRI) provides a high signal-to-noise ratio (SNR), enabling exceptional spatial resolution for clinical diagnostics and research. However, higher fields introduce challenges such as transmit radiofrequency (RF) field inhomogeneities, which result in uneven flip angles and image intensity artifacts. These artifacts degrade image quality and limit clinical adoption. Traditional RF shimming methods, including Magnitude Least Squares (MLS) optimization, mitigate RF field inhomogeneity but are time-intensive and often require the presence of the patient. Recent machine learning methods, such as RF Shim Prediction by Iteratively Projected Ridge Regression and other deep learning architectures, offer alternative approaches but face challenges such as extensive training requirements, limited complexity, and practical data constraints. This paper introduces a holistic learning-based framework called Fast RF Shimming, which achieves a 5000-fold speedup compared to MLS methods. First, random-initialized Adaptive Moment Estimation (Adam) derives reference shimming weights from multichannel RF fields. Next, a Residual Network (ResNet) maps RF fields to shimming outputs while incorporating a confidence parameter into the loss function. Finally, a Non-uniformity Field Detector (NFD) identifies extreme non-uniform outcomes. Comparative evaluations demonstrate significant improvements in both speed and predictive accuracy. The proposed pipeline also supports potential extensions, such as the integration of anatomical priors or multi-echo data, to enhance the robustness of RF field correction. This approach offers a faster and more efficient solution to RF shimming challenges in UHF MRI.
2501.12162
AdaServe: SLO-Customized LLM Serving with Fine-Grained Speculative Decoding
cs.CL cs.AI cs.DC cs.LG
This paper introduces AdaServe, the first LLM serving system to support SLO customization through fine-grained speculative decoding. AdaServe leverages the logits of a draft model to predict the speculative accuracy of tokens and employs a theoretically optimal algorithm to construct token trees for verification. To accommodate diverse SLO requirements without compromising throughput, AdaServe employs a speculation-and-selection scheme that first constructs candidate token trees for each request and then dynamically selects tokens to meet individual SLO constraints while optimizing throughput. Comprehensive evaluations demonstrate that AdaServe achieves up to 73% higher SLO attainment and 74% higher goodput compared to state-of-the-art systems. These results underscore AdaServe's potential to enhance the efficiency and adaptability of LLM deployments across varied application scenarios.
2501.12166
Beyond Window-Based Detection: A Graph-Centric Framework for Discrete Log Anomaly Detection
cs.SE cs.LG
Detecting anomalies in discrete event logs is critical for ensuring system reliability, security, and efficiency. Traditional window-based methods for log anomaly detection often suffer from context bias and fuzzy localization, which hinder their ability to precisely and efficiently identify anomalies. To address these challenges, we propose a graph-centric framework, TempoLog, which leverages multi-scale temporal graph networks for discrete log anomaly detection. Unlike conventional methods, TempoLog constructs continuous-time dynamic graphs directly from event logs, eliminating the need for fixed-size window grouping. By representing log templates as nodes and their temporal relationships as edges, the framework dynamically captures both local and global dependencies across multiple temporal scales. Additionally, a semantic-aware model enhances detection by incorporating rich contextual information. Extensive experiments on public datasets demonstrate that our method achieves state-of-the-art performance in event-level anomaly detection, significantly outperforming existing approaches in both accuracy and efficiency.
2501.12167
Soft-Decision Decoding for LDPC Code-Based Quantitative Group Testing
cs.IT math.IT
We consider the problem of identifying defective items in a population with non-adaptive quantitative group testing. For this scenario, Mashauri et al. recently proposed a low-density parity-check (LDPC) code-based quantitative group testing scheme with a hard-decision decoding approach (akin to peeling decoding). This scheme outperforms generalized LDPC code-based quantitative group testing schemes in terms of the misdetection rate. In this work, we propose a belief-propagation-based decoder for quantitative group testing with LDPC codes, where the messages being passed are purely soft. Through extensive simulations, we show that the proposed soft-information decoder outperforms the hard-decision decoder Mashauri et al.
2501.12169
SVGS-DSGAT: An IoT-Enabled Innovation in Underwater Robotic Object Detection Technology
cs.CV
With the advancement of Internet of Things (IoT) technology, underwater target detection and tracking have become increasingly important for ocean monitoring and resource management. Existing methods often fall short in handling high-noise and low-contrast images in complex underwater environments, lacking precision and robustness. This paper introduces a novel SVGS-DSGAT model that combines GraphSage, SVAM, and DSGAT modules, enhancing feature extraction and target detection capabilities through graph neural networks and attention mechanisms. The model integrates IoT technology to facilitate real-time data collection and processing, optimizing resource allocation and model responsiveness. Experimental results demonstrate that the SVGS-DSGAT model achieves an mAP of 40.8% on the URPC 2020 dataset and 41.5% on the SeaDronesSee dataset, significantly outperforming existing mainstream models. This IoT-enhanced approach not only excels in high-noise and complex backgrounds but also improves the overall efficiency and scalability of the system. This research provides an effective IoT solution for underwater target detection technology, offering significant practical application value and broad development prospects.
2501.12173
ComposeAnyone: Controllable Layout-to-Human Generation with Decoupled Multimodal Conditions
cs.CV
Building on the success of diffusion models, significant advancements have been made in multimodal image generation tasks. Among these, human image generation has emerged as a promising technique, offering the potential to revolutionize the fashion design process. However, existing methods often focus solely on text-to-image or image reference-based human generation, which fails to satisfy the increasingly sophisticated demands. To address the limitations of flexibility and precision in human generation, we introduce ComposeAnyone, a controllable layout-to-human generation method with decoupled multimodal conditions. Specifically, our method allows decoupled control of any part in hand-drawn human layouts using text or reference images, seamlessly integrating them during the generation process. The hand-drawn layout, which utilizes color-blocked geometric shapes such as ellipses and rectangles, can be easily drawn, offering a more flexible and accessible way to define spatial layouts. Additionally, we introduce the ComposeHuman dataset, which provides decoupled text and reference image annotations for different components of each human image, enabling broader applications in human image generation tasks. Extensive experiments on multiple datasets demonstrate that ComposeAnyone generates human images with better alignment to given layouts, text descriptions, and reference images, showcasing its multi-task capability and controllability.
2501.12174
BiMarker: Enhancing Text Watermark Detection for Large Language Models with Bipolar Watermarks
cs.LG
The rapid growth of Large Language Models (LLMs) raises concerns about distinguishing AI-generated text from human content. Existing watermarking techniques, like \kgw, struggle with low watermark strength and stringent false-positive requirements. Our analysis reveals that current methods rely on coarse estimates of non-watermarked text, limiting watermark detectability. To address this, we propose Bipolar Watermark (\tool), which splits generated text into positive and negative poles, enhancing detection without requiring additional computational resources or knowledge of the prompt. Theoretical analysis and experimental results demonstrate \tool's effectiveness and compatibility with existing optimization techniques, providing a new optimization dimension for watermarking in LLM-generated content.
2501.12175
Less is More: Information Bottleneck Denoised Multimedia Recommendation
cs.IR
Empowered by semantic-rich content information, multimedia recommendation has emerged as a potent personalized technique. Current endeavors center around harnessing multimedia content to refine item representation or uncovering latent item-item structures based on modality similarity. Despite the effectiveness, we posit that these methods are usually suboptimal due to the introduction of irrelevant multimedia features into recommendation tasks. This stems from the fact that generic multimedia feature extractors, while well-designed for domain-specific tasks, can inadvertently introduce task-irrelevant features, leading to potential misguidance of recommenders. In this work, we propose a denoised multimedia recommendation paradigm via the Information Bottleneck principle (IB). Specifically, we propose a novel Information Bottleneck denoised Multimedia Recommendation (IBMRec) model to tackle the irrelevant feature issue. IBMRec removes task-irrelevant features from both feature and item-item structure perspectives, which are implemented by two-level IB learning modules: feature-level (FIB) and graph-level (GIB). In particular, FIB focuses on learning the minimal yet sufficient multimedia features. This is achieved by maximizing the mutual information between multimedia representation and recommendation tasks, while concurrently minimizing it between multimedia representation and pre-trained multimedia features. Furthermore, GIB is designed to learn the robust item-item graph structure, it refines the item-item graph based on preference affinity, then minimizes the mutual information between the original graph and the refined one. Extensive experiments across three benchmarks validate the effectiveness of our proposed model, showcasing high performance, and applicability to various multimedia recommenders.
2501.12176
DataPro -- A Standardized Data Understanding and Processing Procedure: A Case Study of an Eco-Driving Project
cs.IR
A systematic pipeline for data processing and knowledge discovery is essential to extracting knowledge from big data and making recommendations for operational decision-making. The CRISP-DM model is the de-facto standard for developing data-mining projects in practice. However, advancements in data processing technologies require enhancements to this framework. This paper presents the DataPro (a standardized data understanding and processing procedure) model, which extends CRISP-DM and emphasizes the link between data scientists and stakeholders by adding the "technical understanding" and "implementation" phases. Firstly, the "technical understanding" phase aligns business demands with technical requirements, ensuring the technical team's accurate comprehension of business goals. Next, the "implementation" phase focuses on the practical application of developed data science models, ensuring theoretical models are effectively applied in business contexts. Furthermore, clearly defining roles and responsibilities in each phase enhances management and communication among all participants. Afterward, a case study on an eco-driving data science project for fuel efficiency analysis in the Danish public transportation sector illustrates the application of the DataPro model. By following the proposed framework, the project identified key business objectives, translated them into technical requirements, and developed models that provided actionable insights for reducing fuel consumption. Finally, the model is evaluated qualitatively, demonstrating its superiority over other data science procedures.
2501.12178
High-dimensional multimodal uncertainty estimation by manifold alignment:Application to 3D right ventricular strain computations
cs.CV
Confidence in the results is a key ingredient to improve the adoption of machine learning methods by clinicians. Uncertainties on the results have been considered in the literature, but mostly those originating from the learning and processing methods. Uncertainty on the data is hardly challenged, as a single sample is often considered representative enough of each subject included in the analysis. In this paper, we propose a representation learning strategy to estimate local uncertainties on a physiological descriptor (here, myocardial deformation) previously obtained from medical images by different definitions or computations. We first use manifold alignment to match the latent representations associated to different high-dimensional input descriptors. Then, we formulate plausible distributions of latent uncertainties, and finally exploit them to reconstruct uncertainties on the input high-dimensional descriptors. We demonstrate its relevance for the quantification of myocardial deformation (strain) from 3D echocardiographic image sequences of the right ventricle, for which a lack of consensus exists in its definition and which directional component to use. We used a database of 100 control subjects with right ventricle overload, for which different types of strain are available at each point of the right ventricle endocardial surface mesh. Our approach quantifies local uncertainties on myocardial deformation from different descriptors defining this physiological concept. Such uncertainties cannot be directly estimated by local statistics on such descriptors, potentially of heterogeneous types. Beyond this controlled illustrative application, our methodology has the potential to be generalized to many other population analyses considering heterogeneous high-dimensional descriptors.
2501.12183
Extend Adversarial Policy Against Neural Machine Translation via Unknown Token
cs.CL
Generating adversarial examples contributes to mainstream neural machine translation~(NMT) robustness. However, popular adversarial policies are apt for fixed tokenization, hindering its efficacy for common character perturbations involving versatile tokenization. Based on existing adversarial generation via reinforcement learning~(RL), we propose the `DexChar policy' that introduces character perturbations for the existing mainstream adversarial policy based on token substitution. Furthermore, we improve the self-supervised matching that provides feedback in RL to cater to the semantic constraints required during training adversaries. Experiments show that our method is compatible with the scenario where baseline adversaries fail, and can generate high-efficiency adversarial examples for analysis and optimization of the system.
2501.12186
Removal of Small Weight Stopping Sets for Asynchronous Unsourced Multiple Access
cs.IT math.IT
In this paper, we analyze the formation of small stopping sets in joint factor graphs describing a frame-asynchronous two-user transmission. Furthermore, we propose an algorithm to completely avoid small stopping sets in the joint factor graph over the entire range of symbol delays. The error floor caused by those stopping sets is completely mitigated. Our key observation is that, while the order of bits in the codeword is irrelevant in a single-user environment, it turns out to be crucial in the asynchronous, unsourced two-user system. Subsequently, our algorithm finds a reordering of variable nodes (VNs) which avoids the smallest stopping set in the joint graph. We show that further improvements can be achieved when girth optimization of the single-user graphs by progressive edge growth (PEG) is used in combination with our proposed algorithm. Starting with a randomized code construction with optimized degree distribution, our simulation results show that PEG followed by the proposed algorithm can improve the average per user probability of error (PUPE) in a noiseless channel by almost two orders of magnitude for a broad range of frame delays.
2501.12189
MirrorCBO: A consensus-based optimization method in the spirit of mirror descent
math.OC cs.LG
In this work we propose MirrorCBO, a consensus-based optimization (CBO) method which generalizes standard CBO in the same way that mirror descent generalizes gradient descent. For this we apply the CBO methodology to a swarm of dual particles and retain the primal particle positions by applying the inverse of the mirror map, which we parametrize as the subdifferential of a strongly convex function $\phi$. In this way, we combine the advantages of a derivative-free non-convex optimization algorithm with those of mirror descent. As a special case, the method extends CBO to optimization problems with convex constraints. Assuming bounds on the Bregman distance associated to $\phi$, we provide asymptotic convergence results for MirrorCBO with explicit exponential rate. Another key contribution is an exploratory numerical study of this new algorithm across different application settings, focusing on (i) sparsity-inducing optimization, and (ii) constrained optimization, demonstrating the competitive performance of MirrorCBO. We observe empirically that the method can also be used for optimization on (non-convex) submanifolds of Euclidean space, can be adapted to mirrored versions of other recent CBO variants, and that it inherits from mirror descent the capability to select desirable minimizers, like sparse ones. We also include an overview of recent CBO approaches for constrained optimization and compare their performance to MirrorCBO.
2501.12191
A margin-based replacement for cross-entropy loss
cs.LG cs.CV
Cross-entropy (CE) loss is the de-facto standard for training deep neural networks to perform classification. However, CE-trained deep neural networks struggle with robustness and generalisation issues. To alleviate these issues, we propose high error margin (HEM) loss, a variant of multi-class margin loss that overcomes the training issues of other margin-based losses. We evaluate HEM extensively on a range of architectures and datasets. We find that HEM loss is more effective than cross-entropy loss across a wide range of tasks: unknown class rejection, adversarial robustness, learning with imbalanced data, continual learning, and semantic segmentation (a pixel-level classification task). Despite all training hyper-parameters being chosen for CE loss, HEM is inferior to CE only in terms of clean accuracy and this difference is insignificant. We also compare HEM to specialised losses that have previously been proposed to improve performance on specific tasks. LogitNorm, a loss achieving state-of-the-art performance on unknown class rejection, produces similar performance to HEM for this task, but is much poorer for continual learning and semantic segmentation. Logit-adjusted loss, designed for imbalanced data, has superior results to HEM for that task, but performs more poorly on unknown class rejection and semantic segmentation. DICE, a popular loss for semantic segmentation, is inferior to HEM loss on all tasks, including semantic segmentation. Thus, HEM often out-performs specialised losses, and in contrast to them, is a general-purpose replacement for CE loss.
2501.12193
MyDigiTwin: A Privacy-Preserving Framework for Personalized Cardiovascular Risk Prediction and Scenario Exploration
cs.LG cs.HC
Cardiovascular disease (CVD) remains a leading cause of death, and primary prevention through personalized interventions is crucial. This paper introduces MyDigiTwin, a framework that integrates health digital twins with personal health environments to empower patients in exploring personalized health scenarios while ensuring data privacy. MyDigiTwin uses federated learning to train predictive models across distributed datasets without transferring raw data, and a novel data harmonization framework addresses semantic and format inconsistencies in health data. A proof-of-concept demonstrates the feasibility of harmonizing and using cohort data to train privacy-preserving CVD prediction models. This framework offers a scalable solution for proactive, personalized cardiovascular care and sets the stage for future applications in real-world healthcare settings.