Summary of X-light: Cross-city Traffic Signal Control Using Transformer on Transformer As Meta Multi-agent Reinforcement Learner, by Haoyuan Jiang et al.
X-Light: Cross-City Traffic Signal Control Using Transformer on Transformer as Meta Multi-Agent Reinforcement Learner
by Haoyuan Jiang, Ziyue Li, Hua Wei, Xuantang Xiong, Jingqing Ruan, Jiaming Lu, Hangyu Mao, Rui Zhao
First submitted to arxiv on: 18 Apr 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper proposes a novel Transformer-based model for cross-city multi-agent traffic signal control, called X-Light. The authors aim to improve the transferability of reinforcement learning-based approaches across diverse cities by leveraging both local and global information. The proposed TonT model consists of two Transformers: the Lower Transformer aggregates state-action-reward trajectories within a city, while the Upper Transformer learns general decision-making patterns across different cities. This dual-level approach enhances the model’s robustness and transferability, outperforming baseline methods with +7.91% on average and up to +16.3% in some cases. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary In this paper, scientists are trying to make traffic lights work better together by using a new kind of artificial intelligence called reinforcement learning. The problem is that these systems don’t work very well when they’re used in different cities. To solve this issue, the researchers created a special model called X-Light that can learn from experiences in one city and apply what it learned to another city. This helps traffic lights make better decisions and reduce congestion. The new approach is much better than previous methods, with an average improvement of 7.91% and up to 16.3% in some cases. |
Keywords
» Artificial intelligence » Reinforcement learning » Transferability » Transformer
