Adaptive Graph Convolutional Recurrent Network for Traffic Forecasting

Abstract

Modeling complex spatial and temporal correlations in the correlated time series data is indispensable for understanding the traffic dynamics and predicting the future status of an evolving traffic system. Recent works focus on designing complicated graph neural network architectures to capture shared patterns with the help of pre-defined graphs. In this paper, we argue that learning node-specific patterns is essential for traffic forecasting while the pre-defined graph is avoidable. To this end, we propose two adaptive modules for enhancing Graph Convolutional Network (GCN) with new capabilities: 1) a Node Adaptive Parameter Learning (NAPL) module to capture node-specific patterns; 2) a Data Adaptive Graph Generation (DAGG) module to infer the inter-dependencies among different traffic series automatically. We further propose an Adaptive Graph Convolutional Recurrent Network (AGCRN) to capture fine-grained spatial and temporal correlations in traffic series automatically based on the two modules and recurrent networks. Our experiments on two real-world traffic datasets show AGCRN outperforms state-of-the-art by a significant margin without pre-defined graphs about spatial connections.

arXiv commentNeurIPS 2020

References in corpus (11)

• Semi-Supervised Classification with Graph Convolutional Networks
• An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling
• A Recurrent Latent Variable Model for Sequential Data
• Deep Learning on Traffic Prediction: Methods, Analysis and Future Directions
• Conditional Time Series Forecasting with Convolutional Neural Networks
• BRITS: Bidirectional Recurrent Imputation for Time Series
• How to Build a Graph-Based Deep Learning Architecture in Traffic Domain: A Survey
• Spatial-Temporal Transformer Networks for Traffic Flow Forecasting
• Deep Multi-View Spatial-Temporal Network for Taxi Demand Prediction
• GMAN: A Graph Multi-Attention Network for Traffic Prediction
• You May Not Need Order in Time Series Forecasting

Cited by in corpus (30)

• Graph Neural Network for Traffic Forecasting: A Survey
• Deep Learning on Traffic Prediction: Methods, Analysis and Future Directions
• A Comprehensive Survey on Deep Graph Representation Learning
• Pre-training Enhanced Spatial-temporal Graph Neural Network for Multivariate Time Series Forecasting
• Traffic Prediction using Artificial Intelligence: Review of Recent Advances and Emerging Opportunities
• SCINet: Time Series Modeling and Forecasting with Sample Convolution and Interaction
• STDEN: Towards Physics-Guided Neural Networks for Traffic Flow Prediction
• Spatial-Temporal Hypergraph Self-Supervised Learning for Crime Prediction
• Regularized Graph Structure Learning with Semantic Knowledge for Multi-variates Time-Series Forecasting
• Spatio-Temporal Graph Neural Point Process for Traffic Congestion Event Prediction
• Domain Adversarial Spatial-Temporal Network: A Transferable Framework for Short-term Traffic Forecasting across Cities
• Short-term Hourly Streamflow Prediction with Graph Convolutional GRU Networks
• Jointly Contrastive Representation Learning on Road Network and Trajectory
• Z-GCNETs: Time Zigzags at Graph Convolutional Networks for Time Series Forecasting
• Continuous-Time and Multi-Level Graph Representation Learning for Origin-Destination Demand Prediction
• Exploring the Generalizability of Spatio-Temporal Traffic Prediction: Meta-Modeling and an Analytic Framework
• Boosting-GNN: Boosting Algorithm for Graph Networks on Imbalanced Node Classification
• Metapopulation Graph Neural Networks: Deep Metapopulation Epidemic Modeling with Human Mobility
• Joint Estimation and Prediction of City-wide Delivery Demand: A Large Language Model Empowered Graph-based Learning Approach
• Adjusting for Autocorrelated Errors in Neural Networks for Time Series
• Network Level Spatial Temporal Traffic State Forecasting with Hierarchical-Attention-LSTM (HierAttnLSTM)
• Because Every Sensor Is Unique, so Is Every Pair: Handling Dynamicity in Traffic Forecasting
• MemDA: Forecasting Urban Time Series with Memory-based Drift Adaptation
• SGRU: A High-Performance Structured Gated Recurrent Unit for Traffic Flow Prediction
• Spatio-Temporal Joint Graph Convolutional Networks for Traffic Forecasting
• Revisiting Random Forests in a Comparative Evaluation of Graph Convolutional Neural Network Variants for Traffic Prediction
• Prediction-based One-shot Dynamic Parking Pricing
• RegionGCN: Spatial-Heterogeneity-Aware Graph Convolutional Networks
• GSA-Forecaster: Forecasting Graph-Based Time-Dependent Data with Graph Sequence Attention
• STR-GODEs: Spatial-Temporal-Ridership Graph ODEs for Metro Ridership Prediction