An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling

2018/03/04 by Shaojie Bai, J. Zico Kolter, Bai, Shaojie +3 · 3 voices · 4,357 citations
Computer Science · #Artificial intelligence #Artificial neural network #Benchmark (surveying) #Code (set theory) #Computer science #Convolutional neural network #Deep learning #Machine learning #Machine translation #Music and Audio Processing #Natural Language Processing Techniques #Natural language processing #Programming language #Recurrent neural network #Sequence (biology) #Sequence labeling #Set (abstract data type) #Task (project management) #Topic Modeling #cs.AI #cs.CL #cs.LG

paper · pdf · doi:10.48550/arxiv.1803.01271

published in arXiv (Cornell University) (Cornell University)

openalex publication_date 2018/03/04 · arxiv created 2018/04/19 · arxiv updated 2018/04/20 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/01

Abstract

For most deep learning practitioners, sequence modeling is synonymous with recurrent networks. Yet recent results indicate that convolutional architectures can outperform recurrent networks on tasks such as audio synthesis and machine translation. Given a new sequence modeling task or dataset, which architecture should one use? We conduct a systematic evaluation of generic convolutional and recurrent architectures for sequence modeling. The models are evaluated across a broad range of standard tasks that are commonly used to benchmark recurrent networks. Our results indicate that a simple convolutional architecture outperforms canonical recurrent networks such as LSTMs across a diverse range of tasks and datasets, while demonstrating longer effective memory. We conclude that the common association between sequence modeling and recurrent networks should be reconsidered, and convolutional networks should be regarded as a natural starting point for sequence modeling tasks. To assist related work, we have made code available at http://github.com/locuslab/TCN .

Citations

• Gradient-based learning applied to document recognition
• Quasi-Recurrent Neural Networks

Cited by

• DM3D: Dynamic Mamba via Offset-Guided Feature Resampling for Point Cloud Understanding
• Pretraining Recurrent Networks without Recurrence
• Climate-resilient electric vehicle charging infrastructure for sustainable cities: An interpretable causal-ensemble framework for preventive maintenance and low-carbon mobility
• Real-Time Detection of Charge Jumps in Superconducting Qubits with a Convolutional Neural Network
• Twoblock clustering trees with coskewness-based dimension reduction: recovering piecewise multivariate linear regimes
• Deep learning-based prediction of time-resolved adhesive forces in viscoelastic Hertzian contacts
• Visual Semantic Decoding of Electrocorticography from Video Stimuli using End-to-End Deep Learning
• Lightweight Wrappers for Adapting Time Series Foundation Models to Regional Drought Forecasting
• Fourier Geometric Wind Power Forecasting with Numerical Weather Prediction
• EdgeCoInfer: Hierarchical Collaborative Inference for On-Device Multimodal Large Models
• A Framework for Early Sepsis Prediction via Self-Supervised (JEPA) and Federated Representation Learning
• HyBDM: Multi-Scale Hybrid Experts for Time Series Forecasting with Bidirectional Dependency Modeling
• Knowledge-Assisted Multi-Graph Dependency Learning for Multivariate Time Series Anomaly Detection in Multi-Stage Industrial Processes
• Structured Latent Space Modeling over Multi-Scale Temporal Patches for Multivariate Time Series Forecasting
• Beyond Output-Space Calibration: Spectral Evidence Bundling for Selective Reliability Estimation in Time-Series Classification
• Trajectory-Aware Clinical Risk Prediction via Severity-Grounded Knowledge Graphs and Retrieval-Augmented Generation
• Chameleon: A Multiplier-Free Temporal Convolutional Network Accelerator for End-to-End Few-Shot and Continual Learning from Sequential Data
• A new adaptive two-layer model for opinion spread in hypergraphs: parameter sensitivity and estimation
• A Neural Network-Based Real-time Casing Collar Recognition System for Downhole Instruments
• What Matters in Deep Learning for Time Series Forecasting?
• SketchMamba: A Lightweight State-Space Model for Joint Progressive Sketch Classification and Stroke Auto-Completion
• Every Client Is an Environment: Federated De-confounding for Spatio-Temporal Forecasting
• Label-free Industrial Fault Detection via Adversarial Inverse Reinforcement Learning: A System for Run-to-Failure Prognostics
• Inferring Missing Trajectory Data with Temporal Convolutional Networks
• Deep Label-Wise Attentive Temporal Convolutional Networks Improve Medical Coding
• Explainable Multimodal Regression via Information Decomposition
• Deep Generative Spatiotemporal Engression for Probabilistic Forecasting of Epidemics
• Spatiotemporal Feature Alignment and Weighted Fusion in Collaborative Perception Enabled by Network Synchronization and Age of Information
• Dual-Prototype Disentanglement: A Context-Aware Enhancement Framework for Time Series Forecasting
• RIPCN: A Road Impedance Principal Component Network for Probabilistic Traffic Flow Forecasting
• RP-CATE: Recurrent Perceptron-based Channel Attention Transformer Encoder for Industrial Hybrid Modeling
• Few-Shot Specific Emitter Identification via Integrated Complex Variational Mode Decomposition and Spatial Attention Transfer
• Concurrence: A dependence criterion for time series, applied to biological data
• BaRISTA: Brain Scale Informed Spatiotemporal Representation of Human Intracranial Neural Activity
• Assessing Neuromorphic Computing for Fingertip Force Decoding from Electromyography
• Augmented Neural Ordinary Differential Equations for Power System Identification
• In-Context and Few-Shots Learning for Forecasting Time Series Data based on Large Language Models
• Hybrid Quantum-Classical Ensemble Learning for S&P 500 Directional Prediction
• Exploiting Spatiotemporal Properties for Efficient Event-Driven Human Pose Estimation
• Physics-Informed Neural Koopman Machine for Interpretable Longitudinal Personalized Alzheimer's Disease Forecasting
• STELLA: Guiding Large Language Models for Time Series Forecasting with Semantic Abstractions
• Comparison of neural network training strategies for the simulation of dynamical systems
• Reasoning-Aware Multimodal Fusion for Hateful Video Detection
• Cuffless Blood Pressure Estimation from Six Wearable Sensor Modalities in Multi-Motion-State Scenarios
• Neonatal seizure detection from raw multi-channel EEG using a fully convolutional architecture
• Privacy-Preserving Generative Modeling and Clinical Validation of Longitudinal Health Records for Chronic Disease
• Spoken Speech Enhancement using EEG
• A Physics-Informed U-net-LSTM Network for Data-Driven Seismic Response Modeling of Structures
• MSTN: Fast and Efficient Multivariate Time Series Prediction Model
• Causal Feature Selection for Weather-Driven Residential Load Forecasting
• Leveraging Spatiotemporal Graph Neural Networks for Multi-Store Sales Forecasting
• Evaluation of Neural Architectures Trained with Square Loss vs Cross-Entropy in Classification Tasks
• Fast Adaptive Neural Control of Resonant Extraction at Fermilab
• Online Multi-Agent Forecasting with Interpretable Collaborative Graph Neural Network
• Autonomous Docking of Multi-Rotor UAVs on Blimps under the Influence of Wind Gusts
• Deep Learning Analysis of Ions Accelerated at Shocks
• Memory-DD: A Low-Complexity Dendrite-Inspired Neuron for Temporal Prediction Tasks
• Physics-Guided Inductive Spatiotemporal Kriging for PM2.5 with Satellite Gradient Constraints
• Transparent Early ICU Mortality Prediction with Clinical Transformer and Per-Case Modality Attribution
• HiPPO: Recurrent Memory with Optimal Polynomial Projections
• A Quantum Tensor Network-Based Viewpoint for Modeling and Analysis of Time Series Data
• Sequential Deep Learning for Credit Risk Monitoring with Tabular Financial Data
• Assessing Generalization in Deep Reinforcement Learning
• DCCRN: Deep Complex Convolution Recurrent Network for Phase-Aware Speech Enhancement
• Gas-insulated switch-gear mechanical fault detection based on acoustic using feature fused neural network
• VitalBench: A Rigorous Multi-Center Benchmark for Long-Term Vital Sign Prediction in Intraoperative Care
• SurvBench: A Standardised Preprocessing Pipeline for Multi-Modal Electronic Health Record Survival Analysis
• X-IONet: Cross-Platform Inertial Odometry Network with Dual-Stage Attention
• Versatile and Risk-Sensitive Cardiac Diagnosis via Graph-Based ECG Signal Representation
• LAD-BNet: Lag-Aware Dual-Branch Networks for Real-Time Energy Forecasting on Edge Devices
• MARAuder's Map: Motion-Aware Real-time Activity Recognition with Layout-Based Trajectories
• Temporal convolutional and fusional transformer model with Bi-LSTM encoder-decoder for multi-time-window remaining useful life prediction
• Predicting Grain Growth in Polycrystalline Materials Using Deep Learning Time Series Models
• A Hybrid Deep Learning based Carbon Price Forecasting Framework with Structural Breakpoints Detection and Signal Denoising
• UnCLe: Towards Scalable Dynamic Causal Discovery in Non-linear Temporal Systems
• Smartphone User Fingerprinting on Wireless Traffic
• Efficient Linear Attention for Multivariate Time Series Modeling via Entropy Equality
• Applying Time Series Deep Learning Models to Forecast the Growth of Perennial Ryegrass in Ireland
• Functional embeddings enable Aggregation of multi-area SEEG recordings over subjects and sessions
• Discovering EV Charging Site Archetypes Through Few Shot Forecasting: The First U.S.-Wide Study
• Towards a predictive spatio-temporal representation of brain data
• DynaBridge: Dynamic Summary-Guided Cross-Task Multimodal Fusion for DASS-Structured Mental Health Assessment
• AdaBERT: Task-Adaptive BERT Compression with Differentiable Neural Architecture Search
• A Multi-Task Learning Framework for Extracting Drugs and Their Interactions from Drug Labels
• Human-Centered Emotion Recognition in Animated GIFs
• Tdcgan: Temporal Dilated Convolutional Generative Adversarial Network for End-to-end Speech Enhancement
• Spotting insects from satellites: modeling the presence of Culicoides imicola through Deep CNNs
• Online Handwriting Trajectory Reconstruction from Kinematic Sensors using Temporal Convolutional Network
• Deep learning for pedestrians: backpropagation in CNNs
• T-Rep: Representation Learning for Time Series using Time-Embeddings
• Deep learning: new computational modelling techniques for genomics
• Divide and Conquer: A Deep CASA Approach to Talker-independent Monaural Speaker Separation
• Improving EEG based Continuous Speech Recognition
• Machine Learning Featurizations for AI Hacking of Political Systems
• Distance-aware Soft Prompt Guidance for Multimodal Valence-Arousal Estimation
• Temporal Graph Convolutional Networks for Automatic Seizure Detection
• Channel Relationship Prediction with Forget-Update Module for Few-shot Classification
• Exploring the Best Loss Function for DNN-Based Low-latency Speech Enhancement with Temporal Convolutional Networks
• Driver2vec: Driver Identification from Automotive Data
• On the Importance of Attention in Meta-Learning for Few-Shot Text Classification
• TPCN: Temporal Point Cloud Networks for Motion Forecasting
• Voice Activity Detection in presence of background noise using EEG
• Learning a Proposal Classifier for Multiple Object Tracking
• Improving the Gating Mechanism of Recurrent Neural Networks
• Compositionality decomposed: how do neural networks generalise?
• Parallel BiLSTM-Transformer networks for forecasting chaotic dynamics
• Sustainable NARMA-10 Benchmarking for Quantum Reservoir Computing
• SARNet: A Spike-Aware consecutive validation Framework for Accurate Remaining Useful Life Prediction
• Spectral Temporal Graph Neural Network for Multivariate Time-series Forecasting
• Dual-path RNN: efficient long sequence modeling for time-domain single-channel speech separation
• Enhancing WiFi CSI Fingerprinting: A Deep Auxiliary Learning Approach
• Error Adjustment Based on Spatiotemporal Correlation Fusion for Traffic Forecasting
• Audio-Linguistic Embeddings for Spoken Sentences
• Elementary, My Dear Watson: Non-Invasive Neural Keyword Spotting in the LibriBrain Dataset
• Comparison of Deep learning models on time series forecasting : a case study of Dissolved Oxygen Prediction
• InvDec: Inverted Decoder for Multivariate Time Series Forecasting with Separated Temporal and Variate Modeling
• SynTSBench: Rethinking Temporal Pattern Learning in Deep Learning Models for Time Series
• Unsupervised Anomaly Prediction with N-BEATS and Graph Neural Network in Multi-variate Semiconductor Process Time Series
• Tune-In: Training Under Negative Environments with Interference for Attention Networks Simulating Cocktail Party Effect
• RatioWaveNet: A Learnable RDWT Front-End for Robust and Interpretable EEG Motor-Imagery Classification
• CKConv: Continuous Kernel Convolution For Sequential Data
• Online Kernel Dynamic Mode Decomposition for Streaming Time Series Forecasting with Adaptive Windowing
• Online Time Series Forecasting with Theoretical Guarantees
• SqueezeCall: nanopore basecalling using a Squeezeformer network
• Xihe: Scalable Zero-Shot Time Series Learner Via Hierarchical Interleaved Block Attention
• Renaissance of RNNs in Streaming Clinical Time Series: Compact Recurrence Remains Competitive with Transformers
• Long Expressive Memory for Sequence Modeling
• Learning to Recognize Correctly Completed Procedure Steps in Egocentric Assembly Videos through Spatio-Temporal Modeling
• SDGraph: Multi-Level Sketch Representation Learning by Sparse-Dense Graph Architecture
• TFGA-Net: Temporal-Frequency Graph Attention Network for Brain-Controlled Speaker Extraction
• U-Time: A Fully Convolutional Network for Time Series Segmentation Applied to Sleep Staging
• Actor-Enriched Time Series Forecasting of Process Performance
• Vision-LLMs for Spatiotemporal Traffic Forecasting
• Multi-Agent Regime-Conditioned Diffusion (MARCD) for CVaR-Constrained Portfolio Decisions
• Interactive Atmospheric Composition Emulation for Next-Generation Earth System Models
• CauchyNet: Compact and Data-Efficient Learning using Holomorphic Activation Functions
• A Unified Frequency Domain Decomposition Framework for Interpretable and Robust Time Series Forecasting
• Real-time 3D Deep Multi-Camera Tracking
• Bidirectional Time-Frequency Pyramid Network for Enhanced Robust EEG Classification
• Extreme value forecasting using relevance-based data augmentation with deep learning models
• Variability Aware Recursive Neural Network (VARNN): A Residual-Memory Model for Capturing Temporal Deviation in Sequence Regression Modeling
• Cross-dataset Multivariate Time-series Model for Parkinson's Diagnosis via Keyboard Dynamics
• STaTS: Structure-Aware Temporal Sequence Summarization via Statistical Window Merging
• IKNet: Interpretable Stock Price Prediction via Keyword-Guided Integration of News and Technical Indicators
• Spatial-Temporal Block and LSTM Network for Pedestrian Trajectories Prediction
• Beyond Sub-6 GHz: Leveraging mmWave Wi-Fi for Gait-Based Person Identification
• Adjusting for Autocorrelated Errors in Neural Networks for Time Series
• Recurrence-Complete Frame-based Action Models
• TimeGazer: Temporal Modeling of Predictive Gaze Stabilization for AR Interaction
• DPA-Net: A Dual-Path Attention Neural Network for Inferring Glycemic Control Metrics from Self-Monitored Blood Glucose Data
• Low-Latency Neural Inference on an Edge Device for Real-Time Handwriting Recognition from EEG Signals
• Improving Clinical Dataset Condensation with Mode Connectivity-based Trajectory Surrogates
• Forecasting-based Biomedical Time-series Data Synthesis for Open Data and Robust AI
• Glocal Information Bottleneck for Time Series Imputation
• Are Pre-trained Convolutions Better than Pre-trained Transformers?
• Kaleidoscope: An Efficient, Learnable Representation For All Structured Linear Maps
• Developing a Sequential Deep Learning Pipeline to Model Alaskan Permafrost Thaw Under Climate Change
• Enhanced 3D Human Pose Estimation from Videos by using Attention-Based Neural Network with Dilated Convolutions
• Efficiently Generating Correlated Sample Paths from Multi-step Time Series Foundation Models
• Time Series Forecasting With Deep Learning: A Survey
• IR-UWB Radar-Based Contactless Silent Speech Recognition with Attention-Enhanced Temporal Convolutional Networks
• Physics-Informed Learning for Human Whole-Body Kinematics Prediction via Sparse IMUs
• Deep Learning-Based Prediction of Energy Decay Curves from Room Geometry and Material Properties
• HyMaTE: A Hybrid Mamba and Transformer Model for EHR Representation Learning
• Channel, Trend and Periodic-Wise Representation Learning for Multivariate Long-term Time Series Forecasting
• Revisiting Multivariate Time Series Forecasting with Missing Values
• Data-Driven Temperature Modelling of Machine Tools by Neural Networks: A Benchmark
• An Evaluation of Trajectory Prediction Approaches and Notes on the TrajNet Benchmark
• Causal Time Series Generation via Diffusion Models
• Why Attention Fails: The Degeneration of Transformers into MLPs in Time Series Forecasting
• DiffNator: Generating Structured Explanations of Time-Series Differences
• Long-Range Dependence in Financial Markets: Empirical Evidence and Generative Modeling Challenges
• Myosotis: structured computation for attention like layer
• TIMED: Adversarial and Autoregressive Refinement of Diffusion-Based Time Series Generation
• Effective Low-Cost Time-Domain Audio Separation Using Globally Attentive Locally Recurrent Networks
• Heterogeneous-Temporal Graph Convolutional Networks: Make the Community Detection Much Better
• Designing Practical Models for Isolated Word Visual Speech Recognition
• Merchant Category Identification Using Credit Card Transactions
• EPNE: Evolutionary Pattern Preserving Network Embedding
• Machine Learning-Based Performance Evaluation of a Solar-Powered Hydrogen Fuel Cell Hybrid in a Radio-Controlled Electric Vehicle
• Conv-like Scale-Fusion Time Series Transformer: A Multi-Scale Representation for Variable-Length Long Time Series
• TSGym: Design Choices for Deep Multivariate Time-Series Forecasting
• The SAGES Critical View of Safety Challenge: A Global Benchmark for AI-Assisted Surgical Quality Assessment
• Prospective Multi-Graph Cohesion for Multivariate Time Series Anomaly Detection
• Deep Learning Inductive Biases for fMRI Time Series Classification during Resting-state and Movie-watching
• Multi-Channel Differential ASR for Robust Wearer Speech Recognition on Smart Glasses
• Sequence to Point Learning Based on Bidirectional Dilated Residual Network for Non Intrusive Load Monitoring
• NORESQA: A Framework for Speech Quality Assessment using Non-Matching References
• R-Transformer: Recurrent Neural Network Enhanced Transformer
• Motion Estimation for Multi-Object Tracking using KalmanNet with Semantic-Independent Encoding
• LoadCNN: A Low Training Cost Deep Learning Model for Day-Ahead Individual Residential Load Forecasting
• Electric Vehicle Identification from Behind Smart Meter Data
• Trellis Networks for Sequence Modeling
• GTS_Forecaster: a novel deep learning based geodetic time series forecasting toolbox with python
• S²Transformer: Scalable Structured Transformers for Global Station Weather Forecasting
• Sandglasset: A Light Multi-Granularity Self-attentive Network For Time-Domain Speech Separation
• Towards Scalable and Structured Spatiotemporal Forecasting
• Forecasting Russian Equipment Losses Using Time Series and Deep Learning Models
• Real-E: A Foundation Benchmark for Advancing Robust and Generalizable Electricity Forecasting
• Write‐An‐Animation: High‐level Text‐based Animation Editing with Character‐Scene Interaction
• VARMA-Enhanced Transformer for Time Series Forecasting
• Masked multi-head self-attention for causal speech enhancement
• COBRA: Multimodal Sensing Deep Learning Framework for Remote Chronic Obesity Management via Wrist-Worn Activity Monitoring
• Improved memory in recurrent neural networks with sequential non-normal dynamics
• AR-KAN: Autoregressive-Weight-Enhanced Kolmogorov-Arnold Network for Time Series Forecasting
• Under Pressure: Learning to Detect Slip with Barometric Tactile Sensors
• Improving noise robust automatic speech recognition with single-channel time-domain enhancement network
• A review of radar-based nowcasting of precipitation and applicable machine learning techniques
• Transformer-based Self-supervised Multimodal Representation Learning for Wearable Emotion Recognition
• Continuous Speech Recognition using EEG and Video
• One Size Does Not Fit All: Multi-Scale, Cascaded RNNs for Radar Classification
• Efficiently Modeling Long Sequences with Structured State Spaces
• Neural Rough Differential Equations for Long Time Series
• Mapping high-performance RNNs to in-memory neuromorphic chips
• Stable Recurrent Models
• Generative Foundation Model for Structured and Unstructured Electronic Health Records
• Time Series Based Network Intrusion Detection using MTF-Aided Transformer
• Exploring Over-stationarization in Deep Learning-based Bus/Tram Arrival Time Prediction: Analysis and Non-stationary Effect Recovery
• Audio2Gestures: Generating Diverse Gestures from Speech Audio with Conditional Variational Autoencoders
• Quantum-Optimized Selective State Space Model for Efficient Time Series Prediction
• Multi-Ontology Integration with Dual-Axis Propagation for Medical Concept Representation
• Deep Generative Modelling of Human Reach-and-Place Action
• Causal Navigation by Continuous-time Neural Networks
• SincQDR-VAD: A Noise-Robust Voice Activity Detection Framework Leveraging Learnable Filters and Ranking-Aware Optimization
• ReLATE+: Unified Framework for Adversarial Attack Detection, Classification, and Resilient Model Selection in Time-Series Classification
• STDiff: A State Transition Diffusion Framework for Time Series Imputation in Industrial Systems
• Frozen in Time: Parameter-Efficient Time Series Transformers via Reservoir-Induced Feature Expansion and Fixed Random Dynamics
• IEA: Inner Ensemble Average within a convolutional neural network
• TransLLM: A Unified Multi-Task Foundation Framework for Urban Transportation via Learnable Prompting
• Organ-Agents: Virtual Human Physiology Simulator via LLMs
• Towards Practical Lipreading with Distilled and Efficient Models
• ERIS: An Energy-Guided Feature Disentanglement Framework for Out-of-Distribution Time Series Classification
• Interpreting Time Series Forecasts with LIME and SHAP: A Case Study on the Air Passengers Dataset
• Deep Relevance Ranking Using Enhanced Document-Query Interactions
• Physiology-informed layered sensing for intelligent human-exoskeleton interaction
• Improving Neural Language Models by Segmenting, Attending, and Predicting the Future
• TriForecaster: A Mixture of Experts Framework for Multi-Region Electric Load Forecasting with Tri-dimensional Specialization
• Learning from Multiple Time Series: A Deep Disentangled Approach to Diversified Time Series Forecasting
• Is Attention All What You Need? -- An Empirical Investigation on Convolution-Based Active Memory and Self-Attention
• Spliced Binned-Pareto Distribution for Robust Modeling of Heavy-tailed Time Series
• Deep Determinantal Point Processes
• Comparative Analysis of Time Series Foundation Models for Demographic Forecasting: Enhancing Predictive Accuracy in US Population Dynamics
• Beyond Constant Parameters: Hyper Prediction Models and HyperMPC
• Effective Sample Size and Generalization Bounds for Temporal Networks
• EnergyPatchTST: Multi-scale Time Series Transformers with Uncertainty Estimation for Energy Forecasting
• Salt-Rock Creep Deformation Forecasting Using Deep Neural Networks and Analytical Models for Subsurface Energy Storage Applications
• Supervised Dynamic Dimension Reduction with Deep Neural Network
• DeepKoopFormer: A Koopman Enhanced Transformer Based Architecture for Time Series Forecasting
• Detecting and measuring respiratory events in horses during exercise with a microphone: deep learning vs. standard signal processing
• Epi²-Net: Advancing Epidemic Dynamics Forecasting with Physics-Inspired Neural Networks
• User Trajectory Prediction Unifying Global and Local Temporal Information
• InfoSyncNet: Information Synchronization Temporal Convolutional Network for Visual Speech Recognition
• DMSC: Dynamic Multi-Scale Coordination Framework for Time Series Forecasting
• Flow Matching for Probabilistic Learning of Dynamical Systems from Missing or Noisy Data
• Learning Pivoting Manipulation with Force and Vision Feedback Using Optimization-based Demonstrations
• AI-driven Orchestration at Scale: Estimating Service Metrics on National-Wide Testbeds
• MoSSDA: A Semi-Supervised Domain Adaptation Framework for Multivariate Time-Series Classification using Momentum Encoder
• Towards Data-Driven Adaptive Exoskeleton Assistance for Post-stroke Gait
• Space-Time-Separable Graph Convolutional Network for Pose Forecasting
• Are Transformers Effective for Time Series Forecasting?
• Time Series Forecasting Through the Lens of Dynamics
• Randomized Overdrive Neural Networks
• Problems of representation of electrocardiograms in convolutional neural networks
• LiteFat: Lightweight Spatio-Temporal Graph Learning for Real-Time Driver Fatigue Detection
• RRTO: A High-Performance Transparent Offloading System for Model Inference in Mobile Edge Computing
• Measuring Dependencies between Biological Signals with Self-supervision, and its Limitations
• Diffusion Denoiser-Aided Gyrocompassing
• Adaptive Fuzzy Time Series Forecasting via Partially Asymmetric Convolution and Sub-Sliding Window Fusion
• Advances and Challenges in Deep Lip Reading
• A Simplified Fully Quantized Transformer for End-to-end Speech Recognition
• Residual Attention Net for Superior Cross-Domain Time Sequence Modeling
• Hybrid semantics-based vulnerability detection incorporating a Temporal Convolutional Network and Self-attention Mechanism
• DEEPF0: End-To-End Fundamental Frequency Estimation for Music and Speech Signals
• Zero-Shot Recommender Systems
• SPADE-S: A Sparsity-Robust Foundational Forecaster
• Beyond exploding and vanishing gradients: analysing RNN training using attractors and smoothness
• 3D Lip Event Detection via Interframe Motion Divergence at Multiple Temporal Resolutions
• Popularity Prediction on Social Platforms with Coupled Graph Neural Networks
• Continuous Emotion Recognition with Audio-visual Leader-follower Attentive Fusion
• A novel temporal convolutional network with residual self-attention mechanism for remaining useful life prediction of rolling bearings
• Deep Learning Based EDM Subgenre Classification using Mel-Spectrogram and Tempogram Features
• Efficient Integration of Multi-channel Information for Speaker-independent Speech Separation
• Instance Explainable Temporal Network For Multivariate Timeseries
• A Token-wise CNN-based Method for Sentence Compression
• Speech Gesture Generation from the Trimodal Context of Text, Audio, and Speaker Identity
• Phase Space Reconstruction Network for Lane Intrusion Action Recognition
• Learning from Scratch: Structurally-masked Transformer for Next Generation Lib-free Simulation
• A Neural Lip-Sync Framework for Synthesizing Photorealistic Virtual News Anchors
• World Model-Based End-to-End Scene Generation for Accident Anticipation in Autonomous Driving
• Fremer: Lightweight and Effective Frequency Transformer for Workload Forecasting in Cloud Services
• A Seq-to-Seq Transformer Premised Temporal Convolutional Network for Chinese Word Segmentation
• Stress Softening Damage in Strongly Nonlinear Viscoelastic Soft Materials A Physics Informed Data Driven Constitutive Model with Time Temperature Coupling
• NWT: Towards natural audio-to-video generation with representation learning
• Recurrent Neural Network from Adder's Perspective: Carry-lookahead RNN
• Offer Personalization using Temporal Convolution Network and Optimization
• An Unsupervised method for OCR Post-Correction and Spelling Normalisation for Finnish
• ICASSP 2021 Deep Noise Suppression Challenge: Decoupling Magnitude and Phase Optimization with a Two-Stage Deep Network
• Recurrent Neural Networks for Time Series Forecasting: Current Status and Future Directions
• Deep Residual-Dense Lattice Network for Speech Enhancement
• Discriminative Multi-modality Speech Recognition
• EEG Foundation Models: A Critical Review of Current Progress and Future Directions
• Aortic Pressure Forecasting with Deep Sequence Learning
• Convolutive Prediction for Monaural Speech Dereverberation and Noisy-Reverberant Speaker Separation
• Why Self-Attention? A Targeted Evaluation of Neural Machine Translation Architectures
• Fourier Basis Mapping: A Time-Frequency Learning Framework for Time Series Forecasting
• Enhancing Generalization in PPG-Based Emotion Measurement with a CNN-TCN-LSTM Model
• Conspiracy to Commit: Information Pollution, Artificial Intelligence, and Real-World Hate Crime
• Short-Term Temporal Convolutional Networks for Dynamic Hand Gesture Recognition
• IML-Spikeformer: Input-aware Multi-Level Spiking Transformer for Speech Processing
• Towards Robust Representation of Limit Orders Books for Deep Learning Models
• Versatile On-device Adaptation at the Edge by Unifying Few-shot, Zero-shot, Continual, and In-context Learning
• ConvDySAT: Deep Neural Representation Learning on Dynamic Graphs via Self-Attention and Convolutional Neural Networks
• Attention-Based Fusion of IQ and FFT Spectrograms with AoA Features for GNSS Jammer Localization
• A multi-task encoder-dual-decoder framework for mixed frequency data prediction
• Mask6D: Masked Pose Priors For 6D Object Pose Estimation
• MCRM: Mother Compact Recurrent Memory
• VIPose: Real-time Visual-Inertial 6D Object Pose Tracking
• A Daily Tourism Demand Prediction Framework Based on Multi-head Attention CNN: The Case of The Foreign Entrant in South Korea
• KnowIt: Deep Time Series Modeling and Interpretation
• How Easy Is It to Learn Motion Models from Widefield Fluorescence Single Particle Tracks?
• A Graph Attention Spatio-temporal Convolutional Network for 3D Human Pose Estimation in Video
• Gait-Based Hand Load Estimation via Deep Latent Variable Models with Auxiliary Information
• MATERO-RCA: Mode-Aware Trajectory-Level Energy-Based Root-Set Optimization for Industrial Root Cause Analysis
• Multimodal Alignment with Cross-Attentive GRUs for Fine-Grained Video Understanding
• A Multi-view Multi-task Learning Framework for Multi-variate Time Series Forecasting
• Self-Attentive Sequential Recommendation
• Continuous Wavelet Transform and Siamese Network-Based Anomaly Detection in Multi-variate Semiconductor Process Time Series
• Unsupervised Representation Learning for Time Series with Temporal Neighborhood Coding
• Evaluation of a Foundational Model and Stochastic Models for Forecasting Sporadic or Spiky Production Outages of High-Performance Machine Learning Services
• Multi-Model Framework for Reconstructing Gamma-Ray Burst Light Curves
• A new machine learning framework for occupational accidents forecasting with safety inspections integration
• When Will It Fail?: Anomaly to Prompt for Forecasting Future Anomalies in Time Series
• Multi-task Offline Reinforcement Learning for Online Advertising in Recommender Systems
• UniCA: Unified Covariate Adaptation for Time Series Foundation Model
• ViFusionTST: Deep Fusion of Time-Series Image Representations from Load Signals for Early Bed-Exit Prediction
• Beyond Static Models: Hypernetworks for Adaptive and Generalizable Forecasting in Complex Parametric Dynamical Systems
• Self-supervised Point Cloud Prediction Using 3D Spatio-temporal Convolutional Networks
• Multi-Task Gaussian Processes and Dilated Convolutional Networks for Reconstruction of Reproductive Hormonal Dynamics
• TEMGNet: Deep Transformer-based Decoding of Upperlimb sEMG for Hand Gestures Recognition
• A novel fusion architecture for detecting Parkinson’s Disease using semi-supervised speech embeddings
• TD3Net: A temporal densely connected multi-dilated convolutional network for lipreading
• Interpretability in Safety-Critical FinancialTrading Systems
• Over-squashing in Spatiotemporal Graph Neural Networks
• Exploring Length Generalization For Transformer-based Speech Enhancement
• Improved Image Reconstruction and Diffusion Parameter Estimation Using a Temporal Convolutional Network Model of Gradient Trajectory Errors
• SKOLR: Structured Koopman Operator Linear RNN for Time-Series Forecasting
• A Gravity-informed Spatiotemporal Transformer for Human Activity Intensity Prediction
• One Person, One Model, One World: Learning Continual User Representation without Forgetting
• Formant Tracking Using Dilated Convolutional Networks Through Dense Connection with Gating Mechanism
• Forecast-Then-Optimize Deep Learning Methods
• 3D human pose estimation with adaptive receptive fields and dilated temporal convolutions
• TCN-DPD: Parameter-Efficient Temporal Convolutional Networks for Wideband Digital Predistortion
• Deep Inertial Pose: A deep learning approach for human pose estimation
• Predicting Video features from EEG and Vice versa
• ECAM: A Contrastive Learning Approach to Avoid Environmental Collision in Trajectory Forecasting
• LETS Forecast: Learning Embedology for Time Series Forecasting
• Bayesian Optimization in Variational Latent Spaces with Dynamic Compression
• Multi-axis Attentive Prediction for Sparse EventData: An Application to Crime Prediction
• Wavelet-based Disentangled Adaptive Normalization for Non-stationary Times Series Forecasting
• Numerical Investigation of Sequence Modeling Theory using Controllable Memory Functions
• A multi-task learning method for extraction of newly constructed areas based on bi-temporal hyperspectral images
• Temporally Folded Convolutional Neural Networks for Sequence Forecasting
• Retrieval Augmented Time Series Forecasting
• Multi-microphone Complex Spectral Mapping for Utterance-wise and Continuous Speech Separation
• Uncovering Insights of Compound Flooding with Data-Driven AI
• Learning a Generative Motion Model from Image Sequences based on a Latent Motion Matrix
• On the Necessity of Multi-Domain Explanation: An Uncertainty Principle Approach for Deep Time Series Models
• Stock Market Telepathy: Graph Neural Networks Predicting the Secret Conversations between MINT and G7 Countries
• Ensemble-Based Peak Demand Probability Density Forecasting with Application to Risk-Aware Power System Scheduling
• Channel-Imposed Fusion: A Simple yet Effective Method for Medical Time Series Classification
• BinConv: A Neural Architecture for Ordinal Encoding in Time-Series Forecasting
• Discovering long term dependencies in noisy time series data using deep learning
• Entanglement for Pattern Learning in Temporal Data with Logarithmic Complexity: Benchmarking on IBM Quantum Hardware
• Using holistic event information in the trigger
• Temporal Convolutional Autoencoder for Interference Mitigation in FMCW Radar Altimeters
• Neuromorphic Sequential Arena: A Benchmark for Neuromorphic Temporal Processing
• Super Characters: A Conversion from Sentiment Classification to Image Classification
• Forecasting Multivariate Urban Data via Decomposition and Spatio-Temporal Graph Analysis
• Understanding Convolutional Neural Networks for Text Classification
• A domain adaptation neural network for digital twin-supported fault diagnosis
• Detecting Informative Channels: ActionFormer
• Chinese Cyberbullying Detection: Dataset, Method, and Validation
• Predicting Parkinson's Disease with Multimodal Irregularly Collected Longitudinal Smartphone Data
• Enhancing Contrastive Learning-based Electrocardiogram Pretrained Model with Patient Memory Queue
• AOL: Adaptive Online Learning for Human Trajectory Prediction in Dynamic Video Scenes
• Towards Machine Learning-based Model Predictive Control for HVAC Control in Multi-Context Buildings at Scale via Ensemble Learning
• BAH Dataset for Ambivalence/Hesitancy Recognition in Videos for Digital Behavioural Change
• In a Silent Way: Communication Between AI and Improvising Musicians\n Beyond Sound
• An Inertial Sequence Learning Framework for Vehicle Speed Estimation via Smartphone IMU
• Spatially Focused Attack against Spatiotemporal Graph Neural Networks
• Learning from Demonstration with Weakly Supervised Disentanglement
• Learning What to Remember: Test-Time Training via Context Distillation
• Mixture of Low Rank Adaptation with Partial Parameter Sharing for Time Series Forecasting
• Wavelet Probabilistic Recurrent Convolutional Network for Multivariate Time Series Classification
• Improving Deep Learning Models via Constraint-Based Domain Knowledge: a Brief Survey
• Beyond All-to-All: Causal-Aligned Transformer with Dynamic Structure Learning for Multivariate Time Series Forecasting
• Rethinking Full Connectivity in Recurrent Neural Networks
• Mini-Game Lifetime Value Prediction in WeChat
• GraphTCN: Spatio-Temporal Interaction Modeling for Human Trajectory Prediction
• Hierarchical Temporal Convolutional Networks for Dynamic Recommender Systems
• Measuring Depression Symptom Severity from Spoken Language and 3D Facial Expressions
• CRAFT: Time Series Forecasting with Cross-Future Behavior Awareness
• Time to Embed: Unlocking Foundation Models for Time Series with Channel Descriptions
• A Graph Attention Based Approach for Trajectory Prediction in Multi-agent Sports Games
• Modeling Electrical Motor Dynamics using Encoder-Decoder with Recurrent Skip Connection
• CATS: Clustering-Aggregated and Time Series for Business Customer Purchase Intention Prediction
• Demystifying When and Why VLAs Fail in Contact-Rich Tasks and How to Fix Them
• PPTNet: A Hybrid Periodic Pattern-Transformer Architecture for Traffic Flow Prediction and Congestion Identification
• LLM-Enhanced Feature Engineering for Multi-Factor Electricity Price Predictions
• Learning IMU Bias with Diffusion Model
• Deep generative modelling of aircraft trajectories in terminal maneuvering areas
• WiFuse: An Attention Mechanism for Human Activity Recognition using Fused CSI Amplitude and Delay-Doppler Channel Features
• Context-Aware Probabilistic Modeling with LLM for Multimodal Time Series Forecasting
• OLMA: One Loss for More Accurate Time Series Forecasting
• MONAQ: Multi-Objective Neural Architecture Querying for Time-Series Analysis on Resource-Constrained Devices
• Probabilistic Motion Modeling from Medical Image Sequences: Application to Cardiac Cine-MRI
• Understanding effect of speech perception in EEG based speech recognition systems
• Predicting Different Acoustic Features from EEG and towards direct synthesis of Audio Waveform from EEG
• Block-Biased Mamba for Long-Range Sequence Processing
• SGCN:Sparse Graph Convolution Network for Pedestrian Trajectory Prediction
• EnvShip: A Unified Framework for Context-Aware and Cross-Region Vessel Trajectory Forecasting
• Towards end-to-end pulsed eddy current classification and regression with CNN
• Comparison of Syntactic and Semantic Representations of Programs in Neural Embeddings
• SocialGrid: A TCN-enhanced Method for Online Discussion Forecasting
• Cluster-and-Conquer: A Framework For Time-Series Forecasting
• Hierarchically Regularized Deep Forecasting
• Sparse Latent Factor Forecaster (SLFF) with Iterative Inference for Transparent Multi-Horizon Commodity Futures Prediction
• Probabilistic Forecasting with Temporal Convolutional Neural Network
• A Novel Framework for Significant Wave Height Prediction based on Adaptive Feature Extraction Time-Frequency Network
• Attention-Enhanced Reservoir Computing as a Multiple Dynamical System Approximator
• Modeling continuous-time stochastic processes using \N-Curve\n mixtures
• Unsupervised Anomaly Detection for Autonomous Robots via Mahalanobis SVDD with Audio-IMU Fusion
• Generative Models for Long Time Series: Approximately Equivariant Recurrent Network Structures for an Adjusted Training Scheme
• Temporal LiDAR Frame Prediction for Autonomous Driving
• Enhancing Black-Litterman Portfolio via Hybrid Forecasting Model Combining Multivariate Decomposition and Noise Reduction
• DeFINE: DEep Factorized INput Token Embeddings for Neural Sequence Modeling
• FlipDial: A Generative Model for Two-Way Visual Dialogue
• Comparing recurrent and convolutional neural networks for predicting wave propagation
• Speaker Identification using EEG
• Instance-wise Graph-based Framework for Multivariate Time Series Forecasting
• Dual-Forecaster: A Multimodal Time Series Model Integrating Descriptive and Predictive Texts
• Dynamic Forecasting and Temporal Feature Evolution of Stock Repurchases in Listed Companies Using Attention-Based Deep Temporal Networks
• Visual Reasoning over Time Series via Multi-Agent System
• Physics Informed Data Driven model for Flood Prediction: Application of Deep Learning in prediction of urban flood development
• Convolutional Neural Network for Trajectory Prediction
• Subtask Gated Networks for Non-Intrusive Load Monitoring
• dynoNet: a neural network architecture for learning dynamical systems
• Expressive Telepresence via Modular Codec Avatars
• Adversarial Interaction Attack: Fooling AI to Misinterpret Human Intentions
• Human Motion Anticipation with Symbolic Label
• Regime-Adaptive Continual Learning for Portfolio Management
• A comparative study of deep learning and ensemble learning to extend the horizon of traffic forecasting
• Mitigating Shared-Private Branch Imbalance via Dual-Branch Rebalancing for Multimodal Sentiment Analysis
• TimeSAE: Causal Sparse Decoding for Faithful Explanations of Black-Box Time Series Models
• AIBuildAI: An AI Agent for Automatically Building AI Models
• Behavior Score Prediction in Resting-State Functional MRI by Deep State Space Modeling
• CalM: A Self-Supervised Foundation Model for Population Dynamics in Calcium Imaging Data
• Indoor Localization Using Smartphone Magnetic with Multi-Scale TCN and LSTM
• Test-Time Adaptation for Non-stationary Time Series: From Synthetic Regime Shifts to Financial Markets
• Beyond Visual Realism: Toward Reliable Financial Time Series Generation
• Forecasting Equity Correlations with Hybrid Transformer Graph Neural Network
• QuantBench: Benchmarking AI Methods for Quantitative Investment
• Fault Detection in New Wind Turbines with Limited Data by Generative Transfer Learning
• Evaluating Time Series Models for Urban Wastewater Management: Predictive Performance, Model Complexity and Resilience
• CANet: ChronoAdaptive Network for Enhanced Long-Term Time Series Forecasting under Non-Stationarity
• Generative Optimization for Incentivized Advertising with Global Level Constraints
• Unifying Physics- and Data-Driven Modeling via Novel Causal Spatiotemporal Graph Neural Network for Interpretable Epidemic Forecasting
• MobiVSR: A Visual Speech Recognition Solution for Mobile Devices
• AegisTS: An Agent-Driven Hierarchical Reinforcement Learning System for Multivariate Time Series Data Cleaning
• Generalization Studies of Neural Network Models for Cardiac Disease Detection Using Limited Channel ECG
• Recurrent Point Review Models
• Autoencoding sensory substitution
• Evaluating Temporal Plasticity in Foundation Time Series Models for Incremental Fine-tuning
• Rethinking Neural Operations for Diverse Tasks
• Real-Time Well Log Prediction From Drilling Data Using Deep Learning
• K2MUSE: A human lower limb multimodal dataset under diverse conditions for facilitating rehabilitation robotics
• Temporal-Framing Adaptive Network for Heart Sound Segmentation Without Prior Knowledge of State Duration
• The Importance of Balanced Data Sets: Analyzing a Vehicle Trajectory Prediction Model based on Neural Networks and Distributed Representations
• A Transferable Adaptive Domain Adversarial Neural Network for Virtual\n Reality Augmented EMG-Based Gesture Recognition
• FurcaNeXt: End-to-end monaural speech separation with dynamic gated dilated temporal convolutional networks
• Model Blending for Text Classification
• WRSE -- a non-parametric weighted-resolution ensemble for predicting\n individual survival distributions in the ICU
• The LSST-DESC 3x2pt Tomography Optimization Challenge
• TRACE: Learned Proprioceptive Odometry for Legged Robots under Unreliable Contact Conditions
• FISformer: Replacing Self-Attention with a Fuzzy Inference System in Transformer Models for Time Series Forecasting
• xLSTM-ECG: Multi-label ECG Classification via Feature Fusion with xLSTM
• Neural Fidelity Calibration for Informative Sim-to-Real Adaptation
• Efficient Traffic State Prediction With Dynamic Joint Spatio-Temporal Relation Inference
• TSP-OCS: A Time-Series Prediction for Optimal Camera Selection in Multi-Viewpoint Surgical Video Analysis
• Probabilistic QoS Metric Forecasting in Delay-Tolerant Networks Using Conditional Diffusion Models on Latent Dynamics
• Fusing Global and Local: Transformer-CNN Synergy for Next-Gen Current Estimation
• Discrete Function Bases and Convolutional Neural Networks

Discussions

• TCNs are better than RNNs [hn, 21 points, 0 comments]
• Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling [hn, 4 points, 0 comments]
• An Empirical Evaluation of Generic CNN and RNN for Sequence Modeling [hn, 2 points, 0 comments]

Related

• Convolutional Sequence to Sequence Learning
• The Importance of the Current Input in Sequence Modeling
• Temporal Convolutional Attention-based Network For Sequence Modeling
• A Critical Review of Recurrent Neural Networks for Sequence Learning
• Structured Sequence Modeling with Graph Convolutional Recurrent Networks