A Black-box Approach for Non-stationary Multi-agent Reinforcement Learning

Published: 16 Jan 2024, Last Modified: 15 Mar 2024ICLR 2024 posterEveryoneRevisionsBibTeX
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.
Keywords: regret analysis, learning in non-stationary games, bandit feedback
Submission Guidelines: I certify that this submission complies with the submission instructions as described on .
TL;DR: We propose a black-box no-regret algorithm applicable for various problems in multi-agent reinforcement learning, including general-sum games, potential games, and Markov games.
Abstract:
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.
Primary Area: learning theory
Submission Number: 6420
Loading