Representation Learning with Contrastive Predictive Coding

Abstract

While supervised learning has enabled great progress in many applications, unsupervised learning has not seen such widespread adoption, and remains an important and challenging endeavor for artificial intelligence. In this work, we propose a universal unsupervised learning approach to extract useful representations from high-dimensional data, which we call Contrastive Predictive Coding. The key insight of our model is to learn such representations by predicting the future in latent space by using powerful autoregressive models. We use a probabilistic contrastive loss which induces the latent space to capture information that is maximally useful to predict future samples. It also makes the model tractable by using negative sampling. While most prior work has focused on evaluating representations for a particular modality, we demonstrate that our approach is able to learn useful representations achieving strong performance on four distinct domains: speech, images, text and reinforcement learning in 3D environments.

Cited by in corpus (472)

• A Simple Framework for Contrastive Learning of Visual Representations
• Shortcut Learning in Deep Neural Networks
• Improved Baselines with Momentum Contrastive Learning
• Unsupervised Learning of Visual Features by Contrasting Cluster Assignments
• WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processing
• Diffusion Models in Vision: A Survey
• Graph Contrastive Learning with Adaptive Augmentation
• Momentum Contrast for Unsupervised Visual Representation Learning
• Data-Efficient Image Recognition with Contrastive Predictive Coding
• Molecular Contrastive Learning of Representations via Graph Neural Networks
• Contrastive Representation Learning: A Framework and Review
• What Makes for Good Views for Contrastive Learning?
• Reproducible scaling laws for contrastive language-image learning
• Improving Graph Collaborative Filtering with Neighborhood-enriched Contrastive Learning
• Contrastive Multiview Coding
• Graph Self-Supervised Learning: A Survey
• Anomaly Detection on Attributed Networks via Contrastive Self-Supervised Learning
• Prototypical Contrastive Learning of Unsupervised Representations
• Big Self-Supervised Models are Strong Semi-Supervised Learners
• Libri-Light: A Benchmark for ASR with Limited or No Supervision
• Self-Supervised Representation Learning: Introduction, Advances and Challenges
• A Survey on Accuracy-oriented Neural Recommendation: From Collaborative Filtering to Information-rich Recommendation
• Rethinking Pre-training and Self-training
• Large Scale Adversarial Representation Learning
• Self-Supervised Speech Representation Learning: A Review
• Learnable latent embeddings for joint behavioral and neural analysis
• Mockingjay: Unsupervised Speech Representation Learning with Deep Bidirectional Transformer Encoders
• Structured Pruning for Deep Convolutional Neural Networks: A survey
• Self-paced Contrastive Learning with Hybrid Memory for Domain Adaptive Object Re-ID
• Using Self-Supervised Learning Can Improve Model Robustness and Uncertainty
• Single-source Domain Expansion Network for Cross-Scene Hyperspectral Image Classification
• TERA: Self-Supervised Learning of Transformer Encoder Representation for Speech
• Contrastive Learning of Subject-Invariant EEG Representations for Cross-Subject Emotion Recognition
• SimGRACE: A Simple Framework for Graph Contrastive Learning without Data Augmentation
• Image Quality Assessment using Contrastive Learning
• TSMixer: Lightweight MLP-Mixer Model for Multivariate Time Series Forecasting
• CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted Instances
• On Mutual Information Maximization for Representation Learning
• Big Bird: Transformers for Longer Sequences
• Out-of-Distribution Generalization via Risk Extrapolation (REx)
• Neural Image Compression for Gigapixel Histopathology Image Analysis
• A Primer on Motion Capture with Deep Learning: Principles, Pitfalls and Perspectives
• Unsupervised Scalable Representation Learning for Multivariate Time Series
• Approximate Nearest Neighbor Negative Contrastive Learning for Dense Text Retrieval
• Learning Video Representations using Contrastive Bidirectional Transformer
• A survey on Semi-, Self- and Unsupervised Learning for Image Classification
• Self-supervised Contrastive Representation Learning for Semi-supervised Time-Series Classification
• Deep Learning for Unsupervised Anomaly Localization in Industrial Images: A Survey
• CURL: Contrastive Unsupervised Representations for Reinforcement Learning
• Vision-Language Models for Medical Report Generation and Visual Question Answering: A Review
• Self-Supervised MultiModal Versatile Networks
• Self-supervised representation learning from 12-lead ECG data
• Momentum Contrastive Learning for Few-Shot COVID-19 Diagnosis from Chest CT Images
• Label-Efficient Self-Supervised Federated Learning for Tackling Data Heterogeneity in Medical Imaging
• Debiased Contrastive Learning for Sequential Recommendation
• Understanding Contrastive Representation Learning through Alignment and Uniformity on the Hypersphere
• Contrastive Masked Autoencoders are Stronger Vision Learners
• A Review-aware Graph Contrastive Learning Framework for Recommendation
• Video Transformers: A Survey
• BigSSL: Exploring the Frontier of Large-Scale Semi-Supervised Learning for Automatic Speech Recognition
• Unsupervised Learning via Meta-Learning
• TCLR: Temporal Contrastive Learning for Video Representation
• Invertible Residual Networks
• Unsupervised Point Cloud Representation Learning with Deep Neural Networks: A Survey
• VLT: Vision-Language Transformer and Query Generation for Referring Segmentation
• Twin Contrastive Learning for Online Clustering
• Self-Supervised Learning for Videos: A Survey
• DeepSMILE: Contrastive self-supervised pre-training benefits MSI and HRD classification directly from H&E whole-slide images in colorectal and breast cancer
• CARLA: Self-supervised Contrastive Representation Learning for Time Series Anomaly Detection
• Dream to Control: Learning Behaviors by Latent Imagination
• SelfHAR: Improving Human Activity Recognition through Self-training with Unlabeled Data
• 3D Self-Supervised Methods for Medical Imaging
• Learning Representation for Clustering via Prototype Scattering and Positive Sampling
• Towards Learning a Universal Non-Semantic Representation of Speech
• Contrastive Code Representation Learning
• Structure-Preserving Image Super-Resolution
• Disentangling Long and Short-Term Interests for Recommendation
• Semantic Models for the First-stage Retrieval: A Comprehensive Review
• Space-Time Correspondence as a Contrastive Random Walk
• Supervised Contrastive Learning
• A Comparison of Discrete and Soft Speech Units for Improved Voice Conversion
• Contrastive Cross-Domain Sequential Recommendation
• Knowledge Enhancement for Contrastive Multi-Behavior Recommendation
• CrossCBR: Cross-view Contrastive Learning for Bundle Recommendation
• Multi-view Intent Disentangle Graph Networks for Bundle Recommendation
• Self-supervised remote sensing feature learning: Learning Paradigms, Challenges, and Future Works
• CMID: A Unified Self-Supervised Learning Framework for Remote Sensing Image Understanding
• CLIP-Art: Contrastive Pre-training for Fine-Grained Art Classification
• A Self-Supervised Gait Encoding Approach with Locality-Awareness for 3D Skeleton Based Person Re-Identification
• Effectiveness of self-supervised pre-training for speech recognition
• The Conditional Entropy Bottleneck
• DeCLUTR: Deep Contrastive Learning for Unsupervised Textual Representations
• PointContrast: Unsupervised Pre-training for 3D Point Cloud Understanding
• Data Augmenting Contrastive Learning of Speech Representations in the Time Domain
• Deep Metric Learning-based Image Retrieval System for Chest Radiograph and its Clinical Applications in COVID-19
• COCOA: Cross Modality Contrastive Learning for Sensor Data
• Self-supervised Autoregressive Domain Adaptation for Time Series Data
• Large-scale Unsupervised Semantic Segmentation
• ProphetNet: Predicting Future N-gram for Sequence-to-Sequence Pre-training
• Learning Predictive Representations for Deformable Objects Using Contrastive Estimation
• Contrastive Collaborative Filtering for Cold-Start Item Recommendation
• What makes instance discrimination good for transfer learning?
• Improving Molecular Contrastive Learning via Faulty Negative Mitigation and Decomposed Fragment Contrast
• Self-supervised deep convolutional neural network for chest X-ray classification
• Putting An End to End-to-End: Gradient-Isolated Learning of Representations
• Selfie: Self-supervised Pretraining for Image Embedding
• InfoXLM: An Information-Theoretic Framework for Cross-Lingual Language Model Pre-Training
• Towards Label-efficient Automatic Diagnosis and Analysis: A Comprehensive Survey of Advanced Deep Learning-based Weakly-supervised, Semi-supervised and Self-supervised Techniques in Histopathological Image Analysis
• Learning Invariant Representations for Reinforcement Learning without Reconstruction
• Unsupervised Automatic Speech Recognition: A Review
• GATE: Graph CCA for Temporal SElf-supervised Learning for Label-efficient fMRI Analysis
• HiCMAE: Hierarchical Contrastive Masked Autoencoder for Self-Supervised Audio-Visual Emotion Recognition
• MGDCF: Distance Learning via Markov Graph Diffusion for Neural Collaborative Filtering
• Contrastive Out-of-Distribution Detection for Pretrained Transformers
• Self-supervised learning through the eyes of a child
• On Representation Knowledge Distillation for Graph Neural Networks
• Automated Spatio-Temporal Graph Contrastive Learning
• Stochastic Latent Actor-Critic: Deep Reinforcement Learning with a Latent Variable Model
• Contrastive Representation Distillation
• Offline-Online Associated Camera-Aware Proxies for Unsupervised Person Re-identification
• Hyperspectral Image Classification With Contrastive Graph Convolutional Network
• AstroCLIP: A Cross-Modal Foundation Model for Galaxies
• A Systematic Survey of Regularization and Normalization in GANs
• User-Centric Conversational Recommendation with Multi-Aspect User Modeling
• Contrast-Phys+: Unsupervised and Weakly-supervised Video-based Remote Physiological Measurement via Spatiotemporal Contrast
• Video Representation Learning with Visual Tempo Consistency
• Formal Limitations on the Measurement of Mutual Information
• Modeling Two-Way Selection Preference for Person-Job Fit
• BYOL for Audio: Exploring Pre-trained General-purpose Audio Representations
• Near-Optimal Representation Learning for Hierarchical Reinforcement Learning
• A Survey of Deep Learning for Scientific Discovery
• Meta reinforcement learning as task inference
• Self-supervised Learning for Video Correspondence Flow
• Contrastive Domain Adaptation for Time-Series via Temporal Mixup
• CoMIR: Contrastive Multimodal Image Representation for Registration
• Graph Communal Contrastive Learning
• Symbolic Pregression: Discovering Physical Laws from Distorted Video
• Meta-TTS: Meta-Learning for Few-Shot Speaker Adaptive Text-to-Speech
• Learning Dynamic Belief Graphs to Generalize on Text-Based Games
• Self-Regulated Learning for Egocentric Video Activity Anticipation
• Unsupervised Deep Representation Learning and Few-Shot Classification of PolSAR Images
• Nuclei Segmentation with Point Annotations from Pathology Images via Self-Supervised Learning and Co-Training
• Data-Efficient Reinforcement Learning with Self-Predictive Representations
• Self-Supervised RF Signal Representation Learning for NextG Signal Classification with Deep Learning
• Mid-Level Visual Representations Improve Generalization and Sample Efficiency for Learning Visuomotor Policies
• The Free Energy Principle for Perception and Action: A Deep Learning Perspective
• Unsupervised Contrastive Learning based Transformer for Lung Nodule Detection
• Automatic Shortcut Removal for Self-Supervised Representation Learning
• Whitening for Self-Supervised Representation Learning
• Company-as-Tribe: Company Financial Risk Assessment on Tribe-Style Graph with Hierarchical Graph Neural Networks
• Improving Generalization for Abstract Reasoning Tasks Using Disentangled Feature Representations
• Neural Predictive Belief Representations
• Exploring Adapter-based Transfer Learning for Recommender Systems: Empirical Studies and Practical Insights
• Detecting Outliers with Foreign Patch Interpolation
• Nonlinear ICA Using Auxiliary Variables and Generalized Contrastive Learning
• CGC: Contrastive Graph Clustering for Community Detection and Tracking
• Audio ALBERT: A Lite BERT for Self-supervised Learning of Audio Representation
• LipLearner: Customizable Silent Speech Interactions on Mobile Devices
• Multi-Label Contrastive Learning for Abstract Visual Reasoning
• Multimodal Foundation Models for Material Property Prediction and Discovery
• Exploiting Transformation Invariance and Equivariance for Self-supervised Sound Localisation
• Contrastive Learning with Stronger Augmentations
• VATLM: Visual-Audio-Text Pre-Training with Unified Masked Prediction for Speech Representation Learning
• Unsupervised Outlier Detection using Memory and Contrastive Learning
• Contrastive Self-supervised Neural Architecture Search
• Auto IV: Counterfactual Prediction via Automatic Instrumental Variable Decomposition
• A Noise-Robust Self-supervised Pre-training Model Based Speech Representation Learning for Automatic Speech Recognition
• Deep Semantic-Visual Alignment for Zero-Shot Remote Sensing Image Scene Classification
• Universal Paralinguistic Speech Representations Using Self-Supervised Conformers
• Heterogeneous Graph Contrastive Learning with Meta-path Contexts and Adaptively Weighted Negative Samples
• Real-time Seismic Intensity Prediction using Self-supervised Contrastive GNN for Earthquake Early Warning
• Context-self contrastive pretraining for crop type semantic segmentation
• Understanding the Limitations of Variational Mutual Information Estimators
• SSAMBA: Self-Supervised Audio Representation Learning with Mamba State Space Model
• Self-supervised Contrastive Learning of Multi-view Facial Expressions
• Conditional Generation of Medical Images via Disentangled Adversarial Inference
• Towards Better Domain Adaptation for Self-supervised Models: A Case Study of Child ASR
• When Search Meets Recommendation: Learning Disentangled Search Representation for Recommendation
• SelfFed: Self-Supervised Federated Learning for Data Heterogeneity and Label Scarcity in Medical Images
• Feature-aware Diversified Re-ranking with Disentangled Representations for Relevant Recommendation
• MTS-DVGAN: Anomaly Detection in Cyber-Physical Systems using a Dual Variational Generative Adversarial Network
• CLOCS: Contrastive Learning of Cardiac Signals Across Space, Time, and Patients
• Information Maximization Clustering via Multi-View Self-Labelling
• Introduction to Latent Variable Energy-Based Models: A Path Towards Autonomous Machine Intelligence
• Discrete and continuous representations and processing in deep learning: Looking forward
• BERTphone: Phonetically-Aware Encoder Representations for Utterance-Level Speaker and Language Recognition
• Contrastive Multi-Level Graph Neural Networks for Session-based Recommendation
• ETC: Encoding Long and Structured Inputs in Transformers
• Learning Robotic Navigation from Experience: Principles, Methods, and Recent Results
• Self-Supervised Relational Reasoning for Representation Learning
• Improving out-of-distribution generalization via multi-task self-supervised pretraining
• Low-Resolution Object Recognition with Cross-Resolution Relational Contrastive Distillation
• CoLES: Contrastive Learning for Event Sequences with Self-Supervision
• Learning to Denoise Biomedical Knowledge Graph for Robust Molecular Interaction Prediction
• Improving Word Translation via Two-Stage Contrastive Learning
• Self-supervised language learning from raw audio: Lessons from the Zero Resource Speech Challenge
• False: False Negative Samples Aware Contrastive Learning for Semantic Segmentation of High-Resolution Remote Sensing Image
• Self-Supervised Learning For Few-Shot Image Classification
• Survey on Hand Gesture Recognition from Visual Input
• Unsupervised State Representation Learning in Atari
• Bi-Level Graph Structure Learning for Next POI Recommendation
• A Contrastive Variational Graph Auto-Encoder for Node Clustering
• A Dual-branch Self-supervised Representation Learning Framework for Tumour Segmentation in Whole Slide Images
• ATST: Audio Representation Learning with Teacher-Student Transformer
• Transfer Learning Framework for Low-Resource Text-to-Speech using a Large-Scale Unlabeled Speech Corpus
• X-GOAL: Multiplex Heterogeneous Graph Prototypical Contrastive Learning
• Improving colonoscopy lesion classification using semi-supervised deep learning
• C3-DINO: Joint Contrastive and Non-contrastive Self-Supervised Learning for Speaker Verification
• Hierarchical Autoregressive Image Models with Auxiliary Decoders
• Token-Level Supervised Contrastive Learning for Punctuation Restoration
• Enriched Music Representations with Multiple Cross-modal Contrastive Learning
• Supervised Contrastive Learning based Dual-Mixer Model for Remaining Useful Life Prediction
• Predictive Information Accelerates Learning in RL
• Deep Reinforcement and InfoMax Learning
• Word Segmentation on Discovered Phone Units with Dynamic Programming and Self-Supervised Scoring
• Boosting Semi-supervised Image Segmentation with Global and Local Mutual Information Regularization
• Multimodal Contrastive Learning via Uni-Modal Coding and Cross-Modal Prediction for Multimodal Sentiment Analysis
• A Review of Emerging Research Directions in Abstract Visual Reasoning
• Sample-efficient Reinforcement Learning Representation Learning with Curiosity Contrastive Forward Dynamics Model
• A Comparative Study of Self-supervised Speech Representation Based Voice Conversion
• Towards Graph Self-Supervised Learning with Contrastive Adjusted Zooming
• CLDG: Contrastive Learning on Dynamic Graphs
• CLIP-ReIdent: Contrastive Training for Player Re-Identification
• Graph Element Networks: adaptive, structured computation and memory
• FedDCSR: Federated Cross-domain Sequential Recommendation via Disentangled Representation Learning
• Contrastive prediction strategies for unsupervised segmentation and categorization of phonemes and words
• Dual Contrastive Learning for Spatio-temporal Representation
• Uncovering the Over-smoothing Challenge in Image Super-Resolution: Entropy-based Quantification and Contrastive Optimization
• View-Invariant, Occlusion-Robust Probabilistic Embedding for Human Pose
• LoCo: Local Contrastive Representation Learning
• Conditional Mutual Information Neural Estimator
• Quasi-Dense Similarity Learning for Multiple Object Tracking
• Investigation of Ensemble features of Self-Supervised Pretrained Models for Automatic Speech Recognition
• Contrastive encoder pre-training-based clustered federated learning for heterogeneous data
• Supervised and Self-supervised Pretraining Based COVID-19 Detection Using Acoustic Breathing/Cough/Speech Signals
• Temporal Representation Learning on Monocular Videos for 3D Human Pose Estimation
• Learning Representations for New Sound Classes With Continual Self-Supervised Learning
• CoVR-2: Automatic Data Construction for Composed Video Retrieval
• Kernelized information bottleneck leads to biologically plausible 3-factor Hebbian learning in deep networks
• Towards Nonlinear Disentanglement in Natural Data with Temporal Sparse Coding
• Unsupervised Representation Learning with Future Observation Prediction for Speech Emotion Recognition
• Efficient Transformers with Dynamic Token Pooling
• CCL: Continual Contrastive Learning for LiDAR Place Recognition
• Multi-queue Momentum Contrast for Microvideo-Product Retrieval
• Probing phoneme, language and speaker information in unsupervised speech representations
• ASiT: Local-Global Audio Spectrogram vIsion Transformer for Event Classification
• Self-supervised Graphs for Audio Representation Learning with Limited Labeled Data
• Debiased Contrastive Learning
• PredNet and Predictive Coding: A Critical Review
• Adaptive Deep Kernel Learning
• Integrating Contrastive Learning with Dynamic Models for Reinforcement Learning from Images
• Self-supervised Cross-silo Federated Neural Architecture Search
• Dual-Modal Attention-Enhanced Text-Video Retrieval with Triplet Partial Margin Contrastive Learning
• Generalized Inter-class Loss for Gait Recognition
• Scaling up sign spotting through sign language dictionaries
• Self-supervised Learning of Audio Representations from Audio-Visual Data using Spatial Alignment
• GeoCLR: Georeference Contrastive Learning for Efficient Seafloor Image Interpretation
• CalibrationPhys: Self-supervised Video-based Heart and Respiratory Rate Measurements by Calibrating Between Multiple Cameras
• Text-Free Prosody-Aware Generative Spoken Language Modeling
• Auxiliary Cross-Modal Representation Learning with Triplet Loss Functions for Online Handwriting Recognition
• When Does Self-supervision Improve Few-shot Learning?
• The Power of Selecting Key Blocks with Local Pre-ranking for Long Document Information Retrieval
• Hierarchical Matching and Reasoning for Multi-Query Image Retrieval
• Universum-inspired Supervised Contrastive Learning
• DECAR: Deep Clustering for learning general-purpose Audio Representations
• PHPQ: Pyramid Hybrid Pooling Quantization for Efficient Fine-Grained Image Retrieval
• Towards Mitigating Dimensional Collapse of Representations in Collaborative Filtering
• A Predictive Approach To Enhance Time-Series Forecasting
• Robust Multiview Multimodal Driver Monitoring System Using Masked Multi-Head Self-Attention
• Lip-to-Speech Synthesis for Arbitrary Speakers in the Wild
• Learning Task-relevant Representations for Generalization via Characteristic Functions of Reward Sequence Distributions
• Supervised Contrastive Learning with Hard Negative Samples
• Exploring Active Learning for Semiconductor Defect Segmentation
• Class Gradient Projection For Continual Learning
• Automatic Medical Report Generation: Methods and Applications
• Multi-Scale and Multi-Layer Contrastive Learning for Domain Generalization
• RPT: Toward Transferable Model on Heterogeneous Researcher Data via Pre-Training
• Variational Cross-Graph Reasoning and Adaptive Structured Semantics Learning for Compositional Temporal Grounding
• Bootstrap Representation Learning for Segmentation on Medical Volumes and Sequences
• Variational Bayesian Optimal Experimental Design with Normalizing Flows
• An Unsupervised Information-Theoretic Perceptual Quality Metric
• Modeling Multiple Views via Implicitly Preserving Global Consistency and Local Complementarity
• MultiCapCLIP: Auto-Encoding Prompts for Zero-Shot Multilingual Visual Captioning
• Content-driven Magnitude-Derivative Spectrum Complementary Learning for Hyperspectral Image Classification
• Passive Non-Line-of-Sight Imaging with Light Transport Modulation
• CCLF: A Contrastive-Curiosity-Driven Learning Framework for Sample-Efficient Reinforcement Learning
• Large-vocabulary forensic pathological analyses via prototypical cross-modal contrastive learning
• Multimodal Information Bottleneck for Deep Reinforcement Learning with Multiple Sensors
• Global-Local Bidirectional Reasoning for Unsupervised Representation Learning of 3D Point Clouds
• Hierarchical Contrastive Motion Learning for Video Action Recognition
• Magic dust for cross-lingual adaptation of monolingual wav2vec-2.0
• Learning the Relation between Similarity Loss and Clustering Loss in Self-Supervised Learning
• Speech Separation with Pretrained Frontend to Minimize Domain Mismatch
• On Learning Universal Representations Across Languages
• Contrastive Learning for Debiased Candidate Generation in Large-Scale Recommender Systems
• A Co-training Approach for Noisy Time Series Learning
• Distilling Localization for Self-Supervised Representation Learning
• Self-supervised Learning from a Multi-view Perspective
• Unsupervised data selection for Speech Recognition with contrastive loss ratios
• Deep Learning Methods for Abstract Visual Reasoning: A Survey on Raven's Progressive Matrices
• Relating by Contrasting: A Data-efficient Framework for Multimodal Generative Models
• Ambiguity-Aware and High-Order Relation Learning for Multi-Grained Image-Text Matching
• Defense for Black-box Attacks on Anti-spoofing Models by Self-Supervised Learning
• Federated Self-Supervised Learning in Heterogeneous Settings: Limits of a Baseline Approach on HAR
• How Useful is Self-Supervised Pretraining for Visual Tasks?
• MOCHA: Real-Time Motion Characterization via Context Matching
• PointSmile: Point Self-supervised Learning via Curriculum Mutual Information
• Unsupervised Discovery of Temporal Structure in Noisy Data with Dynamical Components Analysis
• Multi-label Contrastive Predictive Coding
• Unpacking Information Bottlenecks: Unifying Information-Theoretic Objectives in Deep Learning
• Probing the State of the Art: A Critical Look at Visual Representation Evaluation
• Indonesian Automatic Speech Recognition with XLSR-53
• Label-Efficient Semantic Segmentation of LiDAR Point Clouds in Adverse Weather Conditions
• DimCL: Dimensional Contrastive Learning For Improving Self-Supervised Learning
• Heterogeneous Social Event Detection via Hyperbolic Graph Representations
• Noise Contrastive Meta-Learning for Conditional Density Estimation using Kernel Mean Embeddings
• FreRA: A Frequency-Refined Augmentation for Contrastive Learning on Time Series Classification
• Label Informed Contrastive Pretraining for Node Importance Estimation on Knowledge Graphs
• Dreaming: Model-based Reinforcement Learning by Latent Imagination without Reconstruction
• Label-Efficient Self-Supervised Speaker Verification With Information Maximization and Contrastive Learning
• On the Road with 16 Neurons: Mental Imagery with Bio-inspired Deep Neural Networks
• All by Myself: Learning Individualized Competitive Behaviour with a Contrastive Reinforcement Learning optimization
• Federated Learning with Intermediate Representation Regularization
• Learning Discrete Structured Representations by Adversarially Maximizing Mutual Information
• 3DTINC: Time-Equivariant Non-Contrastive Learning for Predicting Disease Progression from Longitudinal OCTs
• Unsupervised Multiplex Graph Learning with Complementary and Consistent Information
• Self-Supervised Learning Aided Class-Incremental Lifelong Learning
• A Novel Semisupervised Contrastive Regression Framework for Forest Inventory Mapping with Multisensor Satellite Data
• Rank-DistiLLM: Closing the Effectiveness Gap Between Cross-Encoders and LLMs for Passage Re-Ranking
• Distribution-based Emotion Recognition in Conversation
• YOLO11-JDE: Fast and Accurate Multi-Object Tracking with Self-Supervised Re-ID
• Improving Weakly Supervised Visual Grounding by Contrastive Knowledge Distillation
• UnShadowNet: Illumination Critic Guided Contrastive Learning For Shadow Removal
• Robust Basket Recommendation via Noise-tolerated Graph Contrastive Learning
• Estimating semantic structure for the VQA answer space
• A Survey of Latent Factor Models in Recommender Systems
• Augmentation-aware Self-supervised Learning with Conditioned Projector
• DCL: Differential Contrastive Learning for Geometry-Aware Depth Synthesis
• Local Contrastive Feature learning for Tabular Data
• Biologically Plausible Training Mechanisms for Self-Supervised Learning in Deep Networks
• Self-Supervision, Remote Sensing and Abstraction: Representation Learning Across 3 Million Locations
• Transformer based unsupervised pre-training for acoustic representation learning
• Video-Guided Curriculum Learning for Spoken Video Grounding
• Camera-aware Style Separation and Contrastive Learning for Unsupervised Person Re-identification
• Learning Actionable Representations with Goal-Conditioned Policies
• IDEA: Increasing Text Diversity via Online Multi-Label Recognition for Vision-Language Pre-training
• A Generic Self-Supervised Framework of Learning Invariant Discriminative Features
• Unsupervised Pre-Training for 3D Leaf Instance Segmentation
• Understanding Hyperbolic Metric Learning through Hard Negative Sampling
• CPMR: Context-Aware Incremental Sequential Recommendation with Pseudo-Multi-Task Learning
• QAInfomax: Learning Robust Question Answering System by Mutual Information Maximization
• TransformCode: A Contrastive Learning Framework for Code Embedding via Subtree Transformation
• Set-Encoder: Permutation-Invariant Inter-Passage Attention for Listwise Passage Re-Ranking with Cross-Encoders
• Neural Methods for Point-wise Dependency Estimation
• ExpPoint-MAE: Better interpretability and performance for self-supervised point cloud transformers
• Self-supervised Video-centralised Transformer for Video Face Clustering
• Finding online neural update rules by learning to remember
• Representation Learning via Consistent Assignment of Views to Clusters
• Auxiliary Tasks Speed Up Learning PointGoal Navigation
• Recycle-and-Distill: Universal Compression Strategy for Transformer-based Speech SSL Models with Attention Map Reusing and Masking Distillation
• Reviewing Evolution of Learning Functions and Semantic Information Measures for Understanding Deep Learning
• Better with Less: A Data-Active Perspective on Pre-Training Graph Neural Networks
• Variational Sequential Optimal Experimental Design using Reinforcement Learning
• A Brief Overview of Unsupervised Neural Speech Representation Learning
• From Multimodal to Unimodal Attention in Transformers using Knowledge Distillation
• Evaluating the Label Efficiency of Contrastive Self-Supervised Learning for Multi-Resolution Satellite Imagery
• Multimedia Generative Script Learning for Task Planning
• KinePose: A temporally optimized inverse kinematics technique for 6DOF human pose estimation with biomechanical constraints
• Unsupervised Subword Modeling Using Autoregressive Pretraining and Cross-Lingual Phone-Aware Modeling
• CTRLStruct: Dialogue Structure Learning for Open-Domain Response Generation
• Telescoping Density-Ratio Estimation
• Prototypical Contrastive Learning through Alignment and Uniformity for Recommendation
• Self-supervised visual learning for analyzing firearms trafficking activities on the Web
• Transferable Representation Learning in Vision-and-Language Navigation
• FESS Loss: Feature-Enhanced Spatial Segmentation Loss for Optimizing Medical Image Analysis
• Are Labels Necessary for Neural Architecture Search?
• Bootstrap Equilibrium and Probabilistic Speaker Representation Learning for Self-supervised Speaker Verification
• DHOG: Deep Hierarchical Object Grouping
• DistillCSE: Distilled Contrastive Learning for Sentence Embeddings
• UVOSAM: A Mask-free Paradigm for Unsupervised Video Object Segmentation via Segment Anything Model
• Agent Modelling under Partial Observability for Deep Reinforcement Learning
• Predictive Coding for Boosting Deep Reinforcement Learning with Sparse Rewards
• Adaptive Margin Contrastive Learning for Ambiguity-aware 3D Semantic Segmentation
• WAVER: Writing-style Agnostic Text-Video Retrieval via Distilling Vision-Language Models Through Open-Vocabulary Knowledge
• Practical and Consistent Estimation of f-Divergences
• Self-Supervised Representation Learning for Nerve Fiber Distribution Patterns in 3D-PLI
• Text2Poster: Laying out Stylized Texts on Retrieved Images
• Compression theory for inhomogeneous systems
• Self-Supervised Representations Improve End-to-End Speech Translation
• A comprehensive study on the prediction reliability of graph neural networks for virtual screening
• A Semantic Generalization of Shannon's Information Theory and Applications
• On the Estimation of Information Measures of Continuous Distributions
• ConditionNET: Learning Preconditions and Effects for Execution Monitoring
• A Novel Transformer-Based Self-Supervised Learning Method to Enhance Photoplethysmogram Signal Artifact Detection
• Induction Network: Audio-Visual Modality Gap-Bridging for Self-Supervised Sound Source Localization
• Self-Supervised Place Recognition by Refining Temporal and Featural Pseudo Labels from Panoramic Data
• Evolutionary Augmentation Policy Optimization for Self-supervised Learning
• Generative Hierarchical Features from Synthesizing Images
• EqCo: Equivalent Rules for Self-supervised Contrastive Learning
• AutoLTS: Automating Cycling Stress Assessment via Contrastive Learning and Spatial Post-processing
• Conditional Mutual information-based Contrastive Loss for Financial Time Series Forecasting
• DP-Parse: Finding Word Boundaries from Raw Speech with an Instance Lexicon
• Supervised Representation Learning towards Generalizable Assembly State Recognition
• Evaluation of self-supervised pre-training for automatic infant movement classification using wearable movement sensors
• Predictively Encoded Graph Convolutional Network for Noise-Robust Skeleton-based Action Recognition
• Unsupervised Representation Learning by Predicting Random Distances
• NeRF-Feat: 6D Object Pose Estimation using Feature Rendering
• Clustering-Based Representation Learning through Output Translation and Its Application to Remote--Sensing Images
• Multiview Representation Learning from Crowdsourced Triplet Comparisons
• Comparison of Self-Supervised Speech Pre-Training Methods on Flemish Dutch
• Edema Estimation From Facial Images Taken Before and After Dialysis via Contrastive Multi-Patient Pre-Training
• On the difficulty of learning and predicting the long-term dynamics of bouncing objects
• Experiments on Anomaly Detection in Autonomous Driving by Forward-Backward Style Transfers
• Contrastive Learning for Self-Supervised Pre-Training of Point Cloud Segmentation Networks With Image Data
• A Multi-level Alignment Training Scheme for Video-and-Language Grounding
• Self-supervised Feature-Gate Coupling for Dynamic Network Pruning
• Deep Contrastive Multiview Network Embedding
• T-Rep: Representation Learning for Time Series using Time-Embeddings
• Building Minimal and Reusable Causal State Abstractions for Reinforcement Learning
• COBE: Contextualized Object Embeddings from Narrated Instructional Video
• Multi-view Contrastive Learning for Online Knowledge Distillation
• Supervision Accelerates Pre-training in Contrastive Semi-Supervised Learning of Visual Representations
• Enhancing Sound Source Localization via False Negative Elimination
• Towards Better Modeling with Missing Data: A Contrastive Learning-based Visual Analytics Perspective
• Inclusive Data Representation in Federated Learning: A Novel Approach Integrating Textual and Visual Prompt
• Visual Task Progress Estimation with Appearance Invariant Embeddings for Robot Control and Planning
• CSLP-AE: A Contrastive Split-Latent Permutation Autoencoder Framework for Zero-Shot Electroencephalography Signal Conversion
• An end-to-end approach for the verification problem: learning the right distance
• Exploring Representation Learning for Small-Footprint Keyword Spotting
• Shuffle & Divide: Contrastive Learning for Long Text
• Self-Supervised Pretraining on Satellite Imagery: a Case Study on Label-Efficient Vehicle Detection
• Audio Match Cutting: Finding and Creating Matching Audio Transitions in Movies and Videos
• Learning Deep Representations by Mutual Information for Person Re-identification
• Decoupling Global and Local Representations via Invertible Generative Flows
• Guided contrastive self-supervised pre-training for automatic speech recognition
• Scale-specific auxiliary multi-task contrastive learning for deep liver vessel segmentation
• ContraCluster: Learning to Classify without Labels by Contrastive Self-Supervision and Prototype-Based Semi-Supervision
• Self-supervised Benchmark Lottery on ImageNet: Do Marginal Improvements Translate to Improvements on Similar Datasets?
• On the robustness of self-supervised representations for multi-view object classification
• Efficient Text Encoders for Labor Market Analysis
• Enhancing Video Music Recommendation with Transformer-Driven Audio-Visual Embeddings
• K-Shot Contrastive Learning of Visual Features with Multiple Instance Augmentations
• CLOP: Video-and-Language Pre-Training with Knowledge Regularizations
• Efficient Training of Self-Supervised Speech Foundation Models on a Compute Budget
• Predictive Coding Networks and Inference Learning: Tutorial and Survey
• Analyzing Regional Organization of the Human Hippocampus in 3D-PLI Using Contrastive Learning and Geometric Unfolding
• Visually Grounded Continual Language Learning with Selective Specialization
• Simultaneous or Sequential Training? How Speech Representations Cooperate in a Multi-Task Self-Supervised Learning System
• Taming Reversible Halftoning via Predictive Luminance
• Temporal Knowledge Consistency for Unsupervised Visual Representation Learning
• No Representation without Transformation
• Contrastive Video Representation Learning via Adversarial Perturbations
• A Novel Information-Theoretic Objective to Disentangle Representations for Fair Classification
• On Compositions of Transformations in Contrastive Self-Supervised Learning
• ScaleNet: An Unsupervised Representation Learning Method for Limited Information
• RUEL: Retrieval-Augmented User Representation with Edge Browser Logs for Sequential Recommendation
• Synthesizing Photorealistic Images with Deep Generative Learning
• Size Matters: Foundation Model for Czech HTML documents
• Self-Supervised Learning of Context-Aware Pitch Prosody Representations
• A Comprehensive Multi-scale Approach for Speech and Dynamics Synchrony in Talking Head Generation
• Label-free Monitoring of Self-Supervised Learning Progress
• Vision and Structured-Language Pretraining for Cross-Modal Food Retrieval
• Input-independent Attention Weights Are Expressive Enough: A Study of Attention in Self-supervised Audio Transformers
• Learning Representations on the Unit Sphere: Investigating Angular Gaussian and von Mises-Fisher Distributions for Online Continual Learning
• On Equivariant and Invariant Learning of Object Landmark Representations
• Contrastive Learning for Authorship Verification
• Longitudinal Self-Supervised Learning
• Synthetic-to-Real Domain Adaptation for Lane Detection
• Masked Autoencoders with Multi-Window Local-Global Attention Are Better Audio Learners
• Revisiting SSL for sound event detection: complementary fusion and adaptive post-processing
• Exploring internal representation of self-supervised networks: few-shot learning abilities and comparison with human semantics and recognition of objects
• GILBO: One Metric to Measure Them All