X-Light: Cross-City Traffic Signal Control Using Transformer on Transformer as Meta Multi-Agent Reinforcement Learner

Ziyue Li, Haoyuan Jiang, Hua Wei, Xuantang Xiong, Jingqing Ruan, Jiaming Lu, Hangyu Mao, Rui Zhao

Track: Highlight

Categories: Transportation

Keywords: Meta Reinforcement Learning, Traffic Signal Control, Transformer

TL;DR: Can a traffic light controlling agent be trained in Cologne but deployed in New York? Yes! We proposed a "Transformer on Transformer" backbone as the meta-reinforcement learner.

Abstract: The effectiveness of traffic light control has been significantly improved by current reinforcement learning-based approaches via better cooperation among multiple traffic lights. However, a persisting issue remains: how to obtain a multi-agent traffic signal control algorithm with remarkable transferability across diverse cities? In this paper, we propose a Transformer on Transformer (TonT) model for cross-city meta multi-agent traffic signal control, named as X-Light: We input the full Markov Decision Process trajectories, and the Lower Transformer aggregates the states, actions, rewards among the target intersection and its neighbors within a city, and the Upper Transformer learns the general decision trajectories across different cities. This dual-level approach bolsters the model's robust generalization and transferability. Notably, when directly transferring to unseen scenarios, ours surpasses all baseline methods with +7.91% on average, and even +16.3% in some cases, yielding the best results. The code is https://github.com/AnonymousID-submission/X-Light.

Submission Number: 14