Quantifying Interaction Level Between Agents Helps Cost-efficient Generalization in Multi-agent Reinforcement Learning

By Yuxin Chen, Chen Tang, Thomas Tian, Chenran Li, Jinning Li, Masayoshi Tomizuka, and Wei Zhan

Reinforcement Learning Journal, vol. 4, 2024, pp. 1950–1964.

Presented at the Reinforcement Learning Conference (RLC), Amherst Massachusetts, August 9–12, 2024.


Download:

Abstract:

Generalization poses a significant challenge in Multi-agent Reinforcement Learning (MARL). The extent to which unseen co-players influence an agent depends on the agent's policy and the specific scenario. A quantitative examination of this relationship sheds light on how to effectively train agents for diverse scenarios. In this study, we present the Level of Influence (LoI), a metric quantifying the interaction intensity among agents within a given scenario and environment. We observe that, generally, a more diverse set of co-play agents during training enhances the generalization performance of the ego agent; however, this improvement varies across distinct scenarios and environments. LoI proves effective in predicting these improvement disparities within specific scenarios. Furthermore, we introduce a LoI-guided resource allocation method tailored to train a set of policies for diverse scenarios under a constrained budget. Our results demonstrate that strategic resource allocation based on LoI can achieve higher performance than uniform allocation under the same computation budget. The code is available at: https://github.com/ThomasChen98/Level-of-Influence.


Citation Information:

Yuxin Chen, Chen Tang, Thomas Tian, Chenran Li, Jinning Li, Masayoshi Tomizuka, and Wei Zhan. "Quantifying Interaction Level Between Agents Helps Cost-efficient Generalization in Multi-agent Reinforcement Learning." Reinforcement Learning Journal, vol. 4, 2024, pp. 1950–1964.

BibTeX:

@article{chen2024quantifying,
    title={Quantifying Interaction Level Between Agents Helps Cost-efficient Generalization in Multi-agent Reinforcement Learning},
    author={Chen, Yuxin and Tang, Chen and Tian, Thomas and Li, Chenran and Li, Jinning and Tomizuka, Masayoshi and Zhan, Wei},
    journal={Reinforcement Learning Journal},
    volume={4},
    pages={1950--1964},
    year={2024}
}