A Distributed Multi-Agent Reinforcement Learning With Graph Decomposition Approach for Large-Scale Adaptive Traffic Signal Control

Keyphrases

• Distributed multi-agent Systems 100%
• Scale Adaptation 100%
• Adaptive Traffic Signal Control 100%
• Decomposition Approach 100%
• Graph Decomposition 100%
• Multi-agent Reinforcement Learning 100%
• Exponentially Weight 40%
• Waiting Queue 20%
• Connected Vehicle Technology 20%
• Signal Scheduling 20%
• Traffic Network 20%
• Double Deep Q-network 20%
• Prioritized Experience Replay 20%
• Queue Size 20%
• Optimal Signals 20%
• Residual Capacity 20%
• Control Model 20%
• Signal Duration 20%
• Optimal Policy 20%

Computer Science

• Multi-Agent Reinforcement Learning 100%
• Traffic Signal 100%
• Use Case 20%
• Residual Capacity 20%
• Traffic Network 20%
• Deep Q-Network 20%

Engineering

• Control Model 25%
• Optimal Policy 25%