An Instance-Dependent Analysis for the Cooperative Multi-Player Multi-Armed Bandit
Abstract
We study the problem of information sharing and cooperation in Multi-Player Multi-Armed bandits. We propose the first algorithm that achieves logarithmic regret for this problem when the collision reward is unknown. Our results are based on two innovations. First, we show that a simple modification to a successive elimination strategy can be used to allow the players to estimate their suboptimality gaps, up to constant factors, in the absence of collisions. Second, we leverage the first result to design a communication protocol that successfully uses the small reward of collisions to coordinate among players, while preserving meaningful instance-dependent logarithmic regret guarantees.
Cite
Text
Pacchiano et al. "An Instance-Dependent Analysis for the Cooperative Multi-Player Multi-Armed Bandit." Proceedings of The 34th International Conference on Algorithmic Learning Theory, 2023.Markdown
[Pacchiano et al. "An Instance-Dependent Analysis for the Cooperative Multi-Player Multi-Armed Bandit." Proceedings of The 34th International Conference on Algorithmic Learning Theory, 2023.](https://mlanthology.org/alt/2023/pacchiano2023alt-instancedependent/)BibTeX
@inproceedings{pacchiano2023alt-instancedependent,
title = {{An Instance-Dependent Analysis for the Cooperative Multi-Player Multi-Armed Bandit}},
author = {Pacchiano, Aldo and Bartlett, Peter and Jordan, Michael},
booktitle = {Proceedings of The 34th International Conference on Algorithmic Learning Theory},
year = {2023},
pages = {1166-1215},
volume = {201},
url = {https://mlanthology.org/alt/2023/pacchiano2023alt-instancedependent/}
}