End-to-end heterogeneous graph neural networks for traffic assignment

Keyphrases

• Traffic Assignment 100%
• Heterogeneous Graph Neural Network 100%
• Neural Network Model 75%
• Prediction Accuracy 50%
• Traffic Flow Analysis 50%
• Highly Accurate 25%
• Numerical Experiments 25%
• Node-based Approach 25%
• Solution Approach 25%
• Loss Function 25%
• Network Topology 25%
• Large-scale Networks 25%
• Convergence Rate 25%
• Promising Solutions 25%
• Training Strategy 25%
• Flow Capacity 25%
• Traffic Pattern 25%
• Result Prediction 25%
• Urban Transportation Network 25%
• Transportation System 25%
• Flow Conservation 25%
• Adaptive Graph 25%
• Link Flow 25%
• Surrogate Model 25%
• Performance Accuracy 25%
• Conservation Principles 25%
• Capacity Ratio 25%
• User Equilibrium 25%
• Complex Traffic 25%
• Overall Loss 25%
• Virtual Link 25%
• Traffic Flow Prediction 25%
• Graph Attention Network 25%
• Conventional Neural Network 25%
• Convergence Prediction 25%
• Flow Conservation Law 25%
• Spatial Traffic 25%

Computer Science

• Graph Neural Network 100%
• Neural Network Model 75%
• Assignment Problem 50%
• Convergence Rate 25%
• Network Topology 25%
• Prediction Accuracy 25%
• Traffic Pattern 25%
• Attention (Machine Learning) 25%
• Conventional Neural Network 25%
• Transportation Network 25%