CMIMP: Effortlessly Achieving Diverse Population Training for Zero-Shot Coordination
Keywords: reinforcement Learning, zero-shot coordination, population-based training
Abstract: Zero-shot coordination has recently become a hot topic in reinforcement learning research recently. It focuses on the generalization ability of agents, requiring them to coordinate well with collaborators that are not seen before without any fine-tuning. Population-based training has been proven to provide good zero-shot coordination performance; nevertheless, existing algorithms exhibit inefficiency, as the training cost scales linearly with the population size. To address this issue, this paper proposes the Conditional Mutual Information Maximized Population (CMIMP), an efficient training framework comprising two key components: a meta-agent that efficiently realizes a population by selectively sharing parameters across agents, and a mutual information regularizer that guarantees population diversity. To empirically validate the effectiveness of CMIMP, this paper evaluates it along with representational frameworks in Hanabi and confirms its superiority.
Primary Area: reinforcement learning
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.
Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 9340
Loading