Preprint
Jul 2026
A Distributionally Robust Multi-agent Reinforcement Learning Framework for Intelligent Intersection Control
An algorithm-agnostic Distributionally Robust MARL framework integrating an adaptive Contextual-Bandit Worst-Case Estimator (CB-WCE) co-evolves with the traffic controllers by dynamically generating adversarial demand mixtures during training, demonstrating the framework's scalability and potential for resilient real-world urban deployment.
Shuwei Pei, Joran Borger, Arda Kosay et al.
· 0 citations