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

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 .

References in corpus (15)

• Sequence to Sequence Learning with Neural Networks
• Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling
• Natural Language Processing (almost) from Scratch
• WaveNet: A Generative Model for Raw Audio
• On the Properties of Neural Machine Translation: Encoder-Decoder Approaches
• Comparative Study of CNN and RNN for Natural Language Processing
• A Simple Way to Initialize Recurrent Networks of Rectified Linear Units
• Modeling Temporal Dependencies in High-Dimensional Sequences: Application to Polyphonic Music Generation and Transcription
• Pointer Sentinel Mixture Models
• A Convolutional Neural Network for Modelling Sentences
• Regularizing and Optimizing LSTM Language Models
• Quasi-Recurrent Neural Networks
• Neural Machine Translation in Linear Time
• Hierarchical Multiscale Recurrent Neural Networks
• Dilated Recurrent Neural Networks

Cited by in corpus (432)

• Conv-TasNet: Surpassing Ideal Time-Frequency Magnitude Masking for Speech Separation
• Time Series Forecasting With Deep Learning: A Survey
• Recurrent Neural Networks for Time Series Forecasting: Current Status and Future Directions
• Deep Audio-Visual Speech Recognition
• Adaptive Graph Convolutional Recurrent Network for Traffic Forecasting
• An Empirical Survey of Data Augmentation for Time Series Classification with Neural Networks
• Deep Learning for Time Series Anomaly Detection: A Survey
• Efficiently Modeling Long Sequences with Structured State Spaces
• Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecasting
• Parallel Spatio-Temporal Attention-Based TCN for Multivariate Time Series Prediction
• Computational bioacoustics with deep learning: a review and roadmap
• Deep Learning for Time Series Forecasting: Tutorial and Literature Survey
• Multi-task Self-Supervised Learning for Human Activity Detection
• Speech Gesture Generation from the Trimodal Context of Text, Audio, and Speaker Identity
• How to Build a Graph-Based Deep Learning Architecture in Traffic Domain: A Survey
• An Evaluation of Anomaly Detection and Diagnosis in Multivariate Time Series
• Graph-Based Deep Learning for Medical Diagnosis and Analysis: Past, Present and Future
• Neural basis expansion analysis with exogenous variables: Forecasting electricity prices with NBEATSx
• SCINet: Time Series Modeling and Forecasting with Sample Convolution and Interaction
• Machine-Learning-Based Diagnostics of EEG Pathology
• Unsupervised Scalable Representation Learning for Multivariate Time Series
• Multi-level Convolutional Autoencoder Networks for Parametric Prediction of Spatio-temporal Dynamics
• Dynamical Variational Autoencoders: A Comprehensive Review
• Towards Explainable Anticancer Compound Sensitivity Prediction via Multimodal Attention-based Convolutional Encoders
• Bayesian-Deep-Learning Estimation of Earthquake Location from Single-Station Observations
• A Comprehensive Overview and Comparative Analysis on Deep Learning Models: CNN, RNN, LSTM, GRU
• HiPPO: Recurrent Memory with Optimal Polynomial Projections
• Neonatal seizure detection from raw multi-channel EEG using a fully convolutional architecture
• Rolling-Unrolling LSTMs for Action Anticipation from First-Person Video
• An Empirical Evaluation of Deep Learning for ICD-9 Code Assignment using MIMIC-III Clinical Notes
• TFAD: A Decomposition Time Series Anomaly Detection Architecture with Time-Frequency Analysis
• Tropical Cyclone Track Forecasting using Fused Deep Learning from Aligned Reanalysis Data
• STConvS2S: Spatiotemporal Convolutional Sequence to Sequence Network for Weather Forecasting
• Supervised and Unsupervised Neural Approaches to Text Readability
• DTAAD: Dual Tcn-Attention Networks for Anomaly Detection in Multivariate Time Series Data
• Assessing Generalization in Deep Reinforcement Learning
• Measuring Depression Symptom Severity from Spoken Language and 3D Facial Expressions
• Think Globally, Act Locally: A Deep Neural Network Approach to High-Dimensional Time Series Forecasting
• Trellis Networks for Sequence Modeling
• Stable Recurrent Models
• Short-Term Electricity-Load Forecasting by Deep Learning: A Comprehensive Survey
• R-Transformer: Recurrent Neural Network Enhanced Transformer
• SageFormer: Series-Aware Framework for Long-term Multivariate Time Series Forecasting
• Benchmarking Deep Learning Interpretability in Time Series Predictions
• Automated interpretation of congenital heart disease from multi-view echocardiograms
• Self-Attentive Sequential Recommendation
• Combining Recurrent, Convolutional, and Continuous-time Models with Linear State-Space Layers
• Fusing Body Posture with Facial Expressions for Joint Recognition of Affect in Child-Robot Interaction
• Ensemble Conformalized Quantile Regression for Probabilistic Time Series Forecasting
• Neural Memory Networks for Seizure Type Classification
• Lip-reading with Densely Connected Temporal Convolutional Networks
• Skeleton-Based Action Segmentation with Multi-Stage Spatial-Temporal Graph Convolutional Neural Networks
• Detection of gravitational-wave signals from binary neutron star mergers using machine learning
• Email Spam Detection Using Hierarchical Attention Hybrid Deep Learning Method
• DeepTrack: Lightweight Deep Learning for Vehicle Path Prediction in Highways
• Intra-domain and cross-domain transfer learning for time series data -- How transferable are the features?
• Unsupervised Deep Learning for IoT Time Series
• A review of radar-based nowcasting of precipitation and applicable machine learning techniques
• A Radar Signal Deinterleaving Method Based on Semantic Segmentation with Neural Network
• Encoding word order in complex embeddings
• An Evaluation of Trajectory Prediction Approaches and Notes on the TrajNet Benchmark
• DCCRN: Deep Complex Convolution Recurrent Network for Phase-Aware Speech Enhancement
• SELD-TCN: Sound Event Localization & Detection via Temporal Convolutional Networks
• Hierarchical Representation Learning in Graph Neural Networks with Node Decimation Pooling
• Real-Time Well Log Prediction From Drilling Data Using Deep Learning
• MGP-AttTCN: An Interpretable Machine Learning Model for the Prediction of Sepsis
• Darts: User-Friendly Modern Machine Learning for Time Series
• Continuous Human Action Recognition for Human-Machine Interaction: A Review
• Bitcoin Transaction Strategy Construction Based on Deep Reinforcement Learning
• A Spatiotemporal Deep Neural Network for Fine-Grained Multi-Horizon Wind Prediction
• TCGAN: Convolutional Generative Adversarial Network for Time Series Classification and Clustering
• Evaluation of Neural Architectures Trained with Square Loss vs Cross-Entropy in Classification Tasks
• Revisiting Long-term Time Series Forecasting: An Investigation on Linear Mapping
• Dual-path RNN: efficient long sequence modeling for time-domain single-channel speech separation
• Beyond exploding and vanishing gradients: analysing RNN training using attractors and smoothness
• Temporal Graph Convolutional Networks for Automatic Seizure Detection
• A neural network multigrid solver for the Navier-Stokes equations
• Lightweight Neural Architecture Search for Temporal Convolutional Networks at the Edge
• Convolutional Neural Networks for Continuous QoE Prediction in Video Streaming Services
• TinyVers: A Tiny Versatile System-on-chip with State-Retentive eMRAM for ML Inference at the Extreme Edge
• Learning physical properties of anomalous random walks using graph neural networks
• Unsupervised classification to improve the quality of a bird song recording dataset
• Knowledge-based Deep Learning for Modeling Chaotic Systems
• W-Transformers : A Wavelet-based Transformer Framework for Univariate Time Series Forecasting
• Fuzzy temporal convolutional neural networks in P300-based Brain-computer interface for smart home interaction
• Exploring the Best Loss Function for DNN-Based Low-latency Speech Enhancement with Temporal Convolutional Networks
• Diverse Human Motion Prediction Guided by Multi-Level Spatial-Temporal Anchors
• Temporal-Framing Adaptive Network for Heart Sound Segmentation without Prior Knowledge of State Duration
• Learning Uncertainty with Artificial Neural Networks for Improved Predictive Process Monitoring
• Why Self-Attention? A Targeted Evaluation of Neural Machine Translation Architectures
• Comparing recurrent and convolutional neural networks for predicting wave propagation
• Are Pre-trained Convolutions Better than Pre-trained Transformers?
• A comparative assessment of deep learning models for day-ahead load forecasting: Investigating key accuracy drivers
• Accuracy of neural networks for the simulation of chaotic dynamics: precision of training data vs precision of the algorithm
• Unsupervised Representation Learning for Time Series with Temporal Neighborhood Coding
• A Lane-Changing Prediction Method Based on Temporal Convolution Network
• Physics Informed Data Driven model for Flood Prediction: Application of Deep Learning in prediction of urban flood development
• Adaptive Dependency Learning Graph Neural Networks
• Characterization of anomalous diffusion through convolutional transformers
• Time Series Analysis and Modeling to Forecast: a Survey
• Short-term daily precipitation forecasting with seasonally-integrated autoencoder
• Improving the Gating Mechanism of Recurrent Neural Networks
• Predicting Like A Pilot: Dataset and Method to Predict Socially-Aware Aircraft Trajectories in Non-Towered Terminal Airspace
• A Novel Temporal Multi-Gate Mixture-of-Experts Approach for Vehicle Trajectory and Driving Intention Prediction
• DSAM: A Deep Learning Framework for Analyzing Temporal and Spatial Dynamics in Brain Networks
• Infant movement classification through pressure distribution analysis
• An Evaluation of Machine Learning and Deep Learning Models for Drought Prediction using Weather Data
• Weakly Supervised Multi-Task Representation Learning for Human Activity Analysis Using Wearables
• Pedestrian Trajectory Prediction Based on Social Interactions Learning With Random Weights
• Deep encoder-decoder hierarchical convolutional neural networks for conjugate heat transfer surrogate modeling
• FlexConv: Continuous Kernel Convolutions with Differentiable Kernel Sizes
• Anomaly Detection using Deep Autoencoders for in-situ Wastewater Systems Monitoring Data
• Neural Rough Differential Equations for Long Time Series
• Analyzing Unaligned Multimodal Sequence via Graph Convolution and Graph Pooling Fusion
• Structural block driven - enhanced convolutional neural representation for relation extraction
• Real-Time Detection of Anomalies in Large-Scale Transient Surveys
• Eat-Radar: Continuous Fine-Grained Intake Gesture Detection Using FMCW Radar and 3D Temporal Convolutional Network with Attention
• Causal Navigation by Continuous-time Neural Networks
• NORESQA: A Framework for Speech Quality Assessment using Non-Matching References
• Improving Deep Learning Models via Constraint-Based Domain Knowledge: a Brief Survey
• AdaBERT: Task-Adaptive BERT Compression with Differentiable Neural Architecture Search
• The LSST-DESC 3x2pt Tomography Optimization Challenge
• Robust and Energy-efficient PPG-based Heart-Rate Monitoring
• Leveraging Deep Learning and Digital Twins to Improve Energy Performance of Buildings
• FastWave: Accelerating Autoregressive Convolutional Neural Networks on FPGA
• Deep Multi-Frame MVDR Filtering for Single-Microphone Speech Enhancement
• Hierarchical Temporal Convolutional Networks for Dynamic Recommender Systems
• Understanding Convolutional Neural Networks for Text Classification
• Explainable artificial intelligence model to predict acute critical illness from electronic health records
• Gramian Angular Fields for leveraging pretrained computer vision models with anomalous diffusion trajectories
• Social-IWSTCNN: A Social Interaction-Weighted Spatio-Temporal Convolutional Neural Network for Pedestrian Trajectory Prediction in Urban Traffic Scenarios
• What is Next when Sequential Prediction Meets Implicitly Hard Interaction?
• In Search of Deep Learning Architectures for Load Forecasting: A Comparative Analysis and the Impact of the Covid-19 Pandemic on Model Performance
• Multi-Variate Time Series Forecasting on Variable Subsets
• MAD: Self-Supervised Masked Anomaly Detection Task for Multivariate Time Series
• Rules of the Road: Predicting Driving Behavior with a Convolutional Model of Semantic Interactions
• Probabilistic Motion Modeling from Medical Image Sequences: Application to Cardiac Cine-MRI
• Subtask Gated Networks for Non-Intrusive Load Monitoring
• TEMGNet: Deep Transformer-based Decoding of Upperlimb sEMG for Hand Gestures Recognition
• A Deep Generative Model of Speech Complex Spectrograms
• Comparison of Deep learning models on time series forecasting : a case study of Dissolved Oxygen Prediction
• Classification of Periodic Variable Stars with Novel Cyclic-Permutation Invariant Neural Networks
• Medical Time Series Classification with Hierarchical Attention-based Temporal Convolutional Networks: A Case Study of Myotonic Dystrophy Diagnosis
• Reduced Order Probabilistic Emulation for Physics-Based Thermosphere Models
• Probabilistic Forecasting with Temporal Convolutional Neural Network
• Graph-enabled Reinforcement Learning for Time Series Forecasting with Adaptive Intelligence
• Deep Efficient Continuous Manifold Learning for Time Series Modeling
• A Simplified Fully Quantized Transformer for End-to-end Speech Recognition
• Adjusting for Autocorrelated Errors in Neural Networks for Time Series
• Spatiotemporal information conversion machine for time-series prediction
• Critically Examining the Claimed Value of Convolutions over User-Item Embedding Maps for Recommender Systems
• SGCN:Sparse Graph Convolution Network for Pedestrian Trajectory Prediction
• Deep Relevance Ranking Using Enhanced Document-Query Interactions
• Empirical study on the efficiency of Spiking Neural Networks with axonal delays, and algorithm-hardware benchmarking
• EmoFace: Audio-driven Emotional 3D Face Animation
• Compositionality decomposed: how do neural networks generalise?
• Learning the Pedestrian-Vehicle Interaction for Pedestrian Trajectory Prediction
• One Person, One Model, One World: Learning Continual User Representation without Forgetting
• Real-time 3D Deep Multi-Camera Tracking
• Vau da muntanialas: Energy-efficient multi-die scalable acceleration of RNN inference
• Monotonic Neural Ordinary Differential Equation: Time-series Forecasting for Cumulative Data
• FlipDial: A Generative Model for Two-Way Visual Dialogue
• Rethinking Full Connectivity in Recurrent Neural Networks
• Zero-Shot Recommender Systems
• Ask "Who", Not "What": Bitcoin Volatility Forecasting with Twitter Data
• DeFINE: DEep Factorized INput Token Embeddings for Neural Sequence Modeling
• Self-supervised Point Cloud Prediction Using 3D Spatio-temporal Convolutional Networks
• SalientSleepNet: Multimodal Salient Wave Detection Network for Sleep Staging
• TCN Mapping Optimization for Ultra-Low Power Time-Series Edge Inference
• Training Robust Deep Models for Time-Series Domain: Novel Algorithms and Theoretical Analysis
• SepAl: Sepsis Alerts On Low Power Wearables With Digital Biomarkers and On-Device Tiny Machine Learning
• SEAnet: A Deep Learning Architecture for Data Series Similarity Search
• CKConv: Continuous Kernel Convolution For Sequential Data
• Human Mesh Recovery from Monocular Images via a Skeleton-disentangled Representation
• On the Importance of Attention in Meta-Learning for Few-Shot Text Classification
• The Importance of Balanced Data Sets: Analyzing a Vehicle Trajectory Prediction Model based on Neural Networks and Distributed Representations
• An Unsupervised method for OCR Post-Correction and Spelling Normalisation for Finnish
• Blood Vessel Geometry Synthesis using Generative Adversarial Networks
• Fast, Diverse and Accurate Image Captioning Guided By Part-of-Speech
• A Graph Attention Spatio-temporal Convolutional Network for 3D Human Pose Estimation in Video
• Pruning In Time (PIT): A Lightweight Network Architecture Optimizer for Temporal Convolutional Networks
• Sequential Deep Learning for Credit Risk Monitoring with Tabular Financial Data
• Driver2vec: Driver Identification from Automotive Data
• Bayesian Optimization in Variational Latent Spaces with Dynamic Compression
• NAC-TCN: Temporal Convolutional Networks with Causal Dilated Neighborhood Attention for Emotion Understanding
• Localization of gravitational waves using machine learning
• Convolutional and Deep Learning based techniques for Time Series Ordinal Classification
• State-Regularized Recurrent Neural Networks
• Attention-based Deep Neural Networks for Battery Discharge Capacity Forecasting
• Improving EEG based Continuous Speech Recognition
• Fast Transient Stability Prediction Using Grid-informed Temporal and Topological Embedding Deep Neural Network
• Time Perception Machine: Temporal Point Processes for the When, Where and What of Activity Prediction
• Structure Preservation for the Deep Neural Network Multigrid Solver
• Multi-task Learning via Adaptation to Similar Tasks for Mortality Prediction of Diverse Rare Diseases
• Earnings Prediction with Deep Learning
• 1D Convolutional Neural Network Models for Sleep Arousal Detection
• A Multi-view Multi-task Learning Framework for Multi-variate Time Series Forecasting
• Learning 3D Human Dynamics from Video
• Designing deep neural networks for driver intention recognition
• Effective Low-Cost Time-Domain Audio Separation Using Globally Attentive Locally Recurrent Networks
• Autoencoding sensory substitution
• Understanding Behavior of Clinical Models under Domain Shifts
• SGRU: A High-Performance Structured Gated Recurrent Unit for Traffic Flow Prediction
• Popularity Prediction on Social Platforms with Coupled Graph Neural Networks
• Residual Attention Net for Superior Cross-Domain Time Sequence Modeling
• Online Prediction-Assisted Safe Reinforcement Learning for Electric Vehicle Charging Station Recommendation in Dynamically Coupled Transportation-Power Systems
• AOL: Adaptive Online Learning for Human Trajectory Prediction in Dynamic Video Scenes
• CT Perfusion is All We Need: 4D CNN Segmentation of Penumbra and Core in Patients With Suspected Ischemic Stroke
• Tropical time series, iterated-sums signatures and quasisymmetric functions
• Follow The Rules: Online Signal Temporal Logic Tree Search for Guided Imitation Learning in Stochastic Domains
• Objective comparison of methods to decode anomalous diffusion
• Machine learning-based multiscale constitutive modelling: Development and application to dual-porosity mass transfer
• Kaleidoscope: An Efficient, Learnable Representation For All Structured Linear Maps
• ICASSP 2021 Deep Noise Suppression Challenge: Decoupling Magnitude and Phase Optimization with a Two-Stage Deep Network
• Learning a Proposal Classifier for Multiple Object Tracking
• A General Framework for Generative Self-supervised Learning in Non-invasive Estimation of Physiological Parameters Using Photoplethysmography
• Spatio-Temporal Joint Graph Convolutional Networks for Traffic Forecasting
• Improving noise robust automatic speech recognition with single-channel time-domain enhancement network
• Fault Detection in New Wind Turbines with Limited Data by Generative Transfer Learning
• Fusing Structure from Motion and Simulation-Augmented Pose Regression from Optical Flow for Challenging Indoor Environments
• Contributing Components of Metabolic Energy Models to Metabolic Cost Estimations in Gait
• Write-a-speaker: Text-based Emotional and Rhythmic Talking-head Generation
• Deep Semi-Supervised Learning for Time Series Classification
• A domain adaptation neural network for digital twin-supported fault diagnosis
• From Known to Unknown: Knowledge-guided Transformer for Time-Series Sales Forecasting in Alibaba
• Frost Prediction Using Machine Learning Methods in Fars Province
• Deep Residual-Dense Lattice Network for Speech Enhancement
• TPCN: Temporal Point Cloud Networks for Motion Forecasting
• Convolutional Neural Network for Trajectory Prediction
• Sequence to Point Learning Based on Bidirectional Dilated Residual Network for Non Intrusive Load Monitoring
• Improved memory in recurrent neural networks with sequential non-normal dynamics
• Tdcgan: Temporal Dilated Convolutional Generative Adversarial Network for End-to-end Speech Enhancement
• Sequence-to-Sequence Load Disaggregation Using Multi-Scale Residual Neural Network
• EPNE: Evolutionary Pattern Preserving Network Embedding
• Proprioceptive External Torque Learning for Floating Base Robot and its Applications to Humanoid Locomotion
• Audio2Gestures: Generating Diverse Gestures from Speech Audio with Conditional Variational Autoencoders
• Hierarchically Regularized Deep Forecasting
• Hypergraph Learning based Recommender System for Anomaly Detection, Control and Optimization
• Emo-DNA: Emotion Decoupling and Alignment Learning for Cross-Corpus Speech Emotion Recognition
• WaveBeat: End-to-end beat and downbeat tracking in the time domain
• Learning Disentangled Representations for Time Series
• PAGAN: Portfolio Analysis with Generative Adversarial Networks
• Human Motion Anticipation with Symbolic Label
• Temporal Convolution Derived Multi-Layered Reservoir Computing
• Towards Practical Lipreading with Distilled and Efficient Models
• Pedestrian Trajectory Prediction using Context-Augmented Transformer Networks
• Deep Attention Fusion Feature for Speech Separation with End-to-End Post-filter Method
• Task Embedding Temporal Convolution Networks for Transfer Learning Problems in Renewable Power Time-Series Forecast
• Benchmarking Deep Sequential Models on Volatility Predictions for Financial Time Series
• Divide and Conquer: A Deep CASA Approach to Talker-independent Monaural Speaker Separation
• Audio-Linguistic Embeddings for Spoken Sentences
• MixSeq: Connecting Macroscopic Time Series Forecasting with Microscopic Time Series Data
• Generalization Studies of Neural Network Models for Cardiac Disease Detection Using Limited Channel ECG
• Dilated Temporal Fully-Convolutional Network for Semantic Segmentation of Motion Capture Data
• Expressive Telepresence via Modular Codec Avatars
• ADF & TransApp: A Transformer-Based Framework for Appliance Detection Using Smart Meter Consumption Series
• Instance-wise Graph-based Framework for Multivariate Time Series Forecasting
• AirObject: A Temporally Evolving Graph Embedding for Object Identification
• The Effectiveness of Discretization in Forecasting: An Empirical Study on Neural Time Series Models
• Spatial-Temporal Block and LSTM Network for Pedestrian Trajectories Prediction
• End-to-End Integration of Speech Separation and Voice Activity Detection for Low-Latency Diarization of Telephone Conversations
• A Graph Attention Based Approach for Trajectory Prediction in Multi-agent Sports Games
• Time Distributed Deep Learning Models for Purely Exogenous Forecasting: Application to Water Table Depth Prediction using Weather Image Time Series
• Hierarchical Text Classification of Urdu News using Deep Neural Network
• SAM: Semi-Active Mechanism for Extensible Continuum Manipulator and Real-time Hysteresis Compensation Control Algorithm
• Comparison of neural network architectures for feature extraction from binary black hole merger waveforms
• Online Multi-horizon Transaction Metric Estimation with Multi-modal Learning in Payment Networks
• On Mini-Batch Training with Varying Length Time Series
• Feasibility of Modeling Orthogonal Frequency-Division Multiplexing Communication Signals with Unsupervised Generative Adversarial Networks
• A deep convolutional neural network that is invariant to time rescaling
• Deep Probabilistic Time Series Forecasting using Augmented Recurrent Input for Dynamic Systems
• Sandglasset: A Light Multi-Granularity Self-attentive Network For Time-Domain Speech Separation
• Randomized Overdrive Neural Networks
• dynoNet: a neural network architecture for learning dynamical systems
• GraphTCN: Spatio-Temporal Interaction Modeling for Human Trajectory Prediction
• Cross-Class Relevance Learning for Temporal Concept Localization
• LoadCNN: A Low Training Cost Deep Learning Model for Day-Ahead Individual Residential Load Forecasting
• Universal Approximation of Input-Output Maps by Temporal Convolutional Nets
• Deep learning for pedestrians: backpropagation in CNNs
• Deep Determinantal Point Processes
• The Newton Scheme for Deep Learning
• Learning from Videos with Deep Convolutional LSTM Networks
• CVaR-based Flight Energy Risk Assessment for Multirotor UAVs using a Deep Energy Model
• FurcaNeXt: End-to-end monaural speech separation with dynamic gated dilated temporal convolutional networks
• SurgeryLSTM: A Time-Aware Neural Model for Accurate and Explainable Length of Stay Prediction After Spine Surgery
• Unsupervised dynamic modeling of medical image transformation
• Short-Term Temporal Convolutional Networks for Dynamic Hand Gesture Recognition
• A Multi-Task Learning Framework for Extracting Drugs and Their Interactions from Drug Labels
• Is Attention All What You Need? -- An Empirical Investigation on Convolution-Based Active Memory and Self-Attention
• Improving Audio Anomalies Recognition Using Temporal Convolutional Attention Network
• A Seq-to-Seq Transformer Premised Temporal Convolutional Network for Chinese Word Segmentation
• Learning behavioral context recognition with multi-stream temporal convolutional networks
• Efficient Integration of Multi-channel Information for Speaker-independent Speech Separation
• Convolutive Prediction for Monaural Speech Dereverberation and Noisy-Reverberant Speaker Separation
• ECG-DelNet: Delineation of Ambulatory Electrocardiograms with Mixed Quality Labeling Using Neural Networks
• TCN-DPD: Parameter-Efficient Temporal Convolutional Networks for Wideband Digital Predistortion
• Discriminative Multi-modality Speech Recognition
• Multi-Task Gaussian Processes and Dilated Convolutional Networks for Reconstruction of Reproductive Hormonal Dynamics
• Stable Invariant Models via Koopman Spectra
• Deep Learning Based EDM Subgenre Classification using Mel-Spectrogram and Tempogram Features
• Stock2Vec: A Hybrid Deep Learning Framework for Stock Market Prediction with Representation Learning and Temporal Convolutional Network
• Super Characters: A Conversion from Sentiment Classification to Image Classification
• Temporal LiDAR Frame Prediction for Autonomous Driving
• Detecting message modification attacks on the CAN bus with Temporal Convolutional Networks
• Heterogeneous-Temporal Graph Convolutional Networks: Make the Community Detection Much Better
• Fully Convolutional Spatio-Temporal Models for Representation Learning in Plasma Science
• One Size Does Not Fit All: Multi-Scale, Cascaded RNNs for Radar Classification
• Detecting Pulsars with Neural Networks: A Proof of Concept
• Merchant Category Identification Using Credit Card Transactions
• Fine-Grained Semantic Segmentation of Motion Capture Data using Dilated Temporal Fully-Convolutional Networks
• Constrained Variational Autoencoder for improving EEG based Speech Recognition Systems
• Towards a predictive spatio-temporal representation of brain data
• Deep Spectro-temporal Artifacts for Detecting Synthesized Speech
• Limited Resource Allocation in a Non-Markovian World: The Case of Maternal and Child Healthcare
• Voice Activity Detection in presence of background noise using EEG
• Continuous Emotion Recognition with Audio-visual Leader-follower Attentive Fusion
• A Neural Lip-Sync Framework for Synthesizing Photorealistic Virtual News Anchors
• T-Rep: Representation Learning for Time Series using Time-Embeddings
• Enhanced 3D Human Pose Estimation from Videos by using Attention-Based Neural Network with Dilated Convolutions
• Visual-information-driven model for crowd simulation using temporal convolutional network
• El-CID: A filter for Gravitational-wave Electromagnetic Counterpart Identification
• Online Multi-Agent Forecasting with Interpretable Collaborative Graph Neural Network
• Rethinking Neural Operations for Diverse Tasks
• Temporal Convolution Networks with Positional Encoding for Evoked Expression Estimation
• Advances and Challenges in Deep Lip Reading
• GMP-AR: Granularity Message Passing and Adaptive Reconciliation for Temporal Hierarchy Forecasting
• A low latency ASR-free end to end spoken language understanding system
• ATCN: Resource-Efficient Processing of Time Series on Edge
• In a Silent Way: Communication Between AI and Improvising Musicians Beyond Sound
• Spatially Focused Attack against Spatiotemporal Graph Neural Networks
• Enhancing Generalization in PPG-Based Emotion Measurement with a CNN-TCN-LSTM Model
• NWT: Towards natural audio-to-video generation with representation learning
• Spliced Binned-Pareto Distribution for Robust Modeling of Heavy-tailed Time Series
• Discovering Predictable Latent Factors for Time Series Forecasting
• Scope and Arbitration in Machine Learning Clinical EEG Classification
• Spotting insects from satellites: modeling the presence of Culicoides imicola through Deep CNNs
• DEEPF0: End-To-End Fundamental Frequency Estimation for Music and Speech Signals
• Mining for meaning: from vision to language through multiple networks consensus
• Improving EEG based continuous speech recognition using GAN
• Recurrent Point Review Models
• MobiVSR: A Visual Speech Recognition Solution for Mobile Devices
• Short-Term Stock Price-Trend Prediction Using Meta-Learning
• AI-driven Orchestration at Scale: Estimating Service Metrics on National-Wide Testbeds
• MCRM: Mother Compact Recurrent Memory
• Learning a Generative Motion Model from Image Sequences based on a Latent Motion Matrix
• IEA: Inner Ensemble Average within a convolutional neural network
• Discrete Function Bases and Convolutional Neural Networks
• Formant Tracking Using Dilated Convolutional Networks Through Dense Connection with Gating Mechanism
• Instance Explainable Temporal Network For Multivariate Timeseries
• Understanding effect of speech perception in EEG based speech recognition systems
• Semantic Representation and Inference for NLP
• Learning from Demonstration with Weakly Supervised Disentanglement
• Predicting the outcome of team movements -- Player time series analysis using fuzzy and deep methods for representation learning
• On the Preservation of Spatio-temporal Information in Machine Learning Applications
• SocialGrid: A TCN-enhanced Method for Online Discussion Forecasting
• Speaker Identification using EEG
• Speech Recognition using EEG signals recorded using dry electrodes
• Continuous Silent Speech Recognition using EEG
• A Token-wise CNN-based Method for Sentence Compression
• Sensor Fusion: Gated Recurrent Fusion to Learn Driving Behavior from Temporal Multimodal Data
• Detecting Patterns of Physiological Response to Hemodynamic Stress via Unsupervised Deep Learning
• Evaluating Time Series Models for Urban Wastewater Management: Predictive Performance, Model Complexity and Resilience
• TD3Net: A temporal densely connected multi-dilated convolutional network for lipreading
• Continuous Speech Recognition using EEG and Video
• Conspiracy to Commit: Information Pollution, Artificial Intelligence, and Real-World Hate Crime
• Recurrent Neural Network from Adder's Perspective: Carry-lookahead RNN
• Mapping high-performance RNNs to in-memory neuromorphic chips
• Peak Alignment of Gas Chromatography-Mass Spectrometry Data with Deep Learning
• Embedding Symbolic Temporal Knowledge into Deep Sequential Models
• Error Adjustment Based on Spatiotemporal Correlation Fusion for Traffic Forecasting
• Early Recognition of Ball Catching Success in Clinical Trials with RNN-Based Predictive Classification
• Transformer-based Self-supervised Multimodal Representation Learning for Wearable Emotion Recognition
• ConvDySAT: Deep Neural Representation Learning on Dynamic Graphs via Self-Attention and Convolutional Neural Networks
• Machine Learning Featurizations for AI Hacking of Political Systems
• A Daily Tourism Demand Prediction Framework Based on Multi-head Attention CNN: The Case of The Foreign Entrant in South Korea
• Long Expressive Memory for Sequence Modeling
• Skeleton-Graph: Long-Term 3D Motion Prediction From 2D Observations Using Deep Spatio-Temporal Graph CNNs
• Predicting Video features from EEG and Vice versa
• WRSE -- a non-parametric weighted-resolution ensemble for predicting individual survival distributions in the ICU
• Indoor Localization Using Smartphone Magnetic with Multi-Scale TCN and LSTM
• Interpretability in Safety-Critical FinancialTrading Systems
• Dynamic Virtual Graph Significance Networks for Predicting Influenza
• Attention Filtering for Multi-person Spatiotemporal Action Detection on Deep Two-Stream CNN Architectures
• REflex: Flexible Framework for Relation Extraction in Multiple Domains
• Multi-microphone Complex Spectral Mapping for Utterance-wise and Continuous Speech Separation
• Modeling Electrical Motor Dynamics using Encoder-Decoder with Recurrent Skip Connection
• Temporally Folded Convolutional Neural Networks for Sequence Forecasting
• Using holistic event information in the trigger
• Offer Personalization using Temporal Convolution Network and Optimization
• Efficient Purely Convolutional Text Encoding
• Graph Deep Factors for Forecasting
• 3D Lip Event Detection via Interframe Motion Divergence at Multiple Temporal Resolutions
• Comparison of Syntactic and Semantic Representations of Programs in Neural Embeddings
• A Hybrid 3DCNN and 3DC-LSTM based model for 4D Spatio-temporal fMRI data: An ABIDE Autism Classification study
• Discovering long term dependencies in noisy time series data using deep learning
• Problems of representation of electrocardiograms in convolutional neural networks
• Predicting Parkinson's Disease with Multimodal Irregularly Collected Longitudinal Smartphone Data
• Adversarial Interaction Attack: Fooling AI to Misinterpret Human Intentions
• Improving Neural Language Models by Segmenting, Attending, and Predicting the Future
• Speaker and Direction Inferred Dual-channel Speech Separation
• CommUnet: U-net decoder for convolutional codes in communication
• Multi-axis Attentive Prediction for Sparse EventData: An Application to Crime Prediction
• Space-Time-Separable Graph Convolutional Network for Pose Forecasting
• Under Pressure: Learning to Detect Slip with Barometric Tactile Sensors
• Cluster-and-Conquer: A Framework For Time-Series Forecasting
• Attentional-GCNN: Adaptive Pedestrian Trajectory Prediction towards Generic Autonomous Vehicle Use Cases
• Dynamic Matching Markets in Power Grid: Concepts and Solution using Deep Reinforcement Learning
• Towards end-to-end pulsed eddy current classification and regression with CNN
• Tune-In: Training Under Negative Environments with Interference for Attention Networks Simulating Cocktail Party Effect
• Learning from Multiple Time Series: A Deep Disentangled Approach to Diversified Time Series Forecasting
• Generalized One-Class Learning Using Pairs of Complementary Classifiers
• Predicting Human Trajectories by Learning and Matching Patterns
• MegazordNet: combining statistical and machine learning standpoints for time series forecasting
• Convolutional Temporal Attention Model for Video-based Person Re-identification
• VIPose: Real-time Visual-Inertial 6D Object Pose Tracking
• A Transferable Adaptive Domain Adversarial Neural Network for Virtual Reality Augmented EMG-Based Gesture Recognition
• Xi-Vector Embedding for Speaker Recognition
• Multi-Modal Temporal Convolutional Network for Anticipating Actions in Egocentric Videos
• Single Model for Influenza Forecasting of Multiple Countries by Multi-task Learning
• Deep Generative Modelling of Human Reach-and-Place Action
• Channel Relationship Prediction with Forget-Update Module for Few-shot Classification
• Impact of Channel Variation on One-Class Learning for Spoof Detection
• Scene Text Recognition with Temporal Convolutional Encoder
• Phase Space Reconstruction Network for Lane Intrusion Action Recognition
• Human-Centered Emotion Recognition in Animated GIFs
• Neural network compression via learnable wavelet transforms
• Modeling continuous-time stochastic processes using $\mathcal{N}$-Curve mixtures
• Aortic Pressure Forecasting with Deep Sequence Learning
• Predicting Different Acoustic Features from EEG and towards direct synthesis of Audio Waveform from EEG
• Living in a Material World: Learning Material Properties from Full-Waveform Flash Lidar Data for Semantic Segmentation
• 3D human pose estimation with adaptive receptive fields and dilated temporal convolutions
• Spoken Speech Enhancement using EEG