English

Multiplayer bandits without observing collision information

Machine Learning 2021-04-06 v2 Computer Science and Game Theory Machine Learning

Abstract

We study multiplayer stochastic multi-armed bandit problems in which the players cannot communicate and if two or more players pull the same arm, a collision occurs and the involved players receive zero reward. We consider two feedback models: a model in which the players can observe whether a collision has occurred and a more difficult setup when no collision information is available. We give the first theoretical guarantees for the second model: an algorithm with a logarithmic regret, and an algorithm with a square-root regret type that does not depend on the gaps between the means. For the first model, we give the first square-root regret bounds that do not depend on the gaps. Building on these ideas, we also give an algorithm for reaching approximate Nash equilibria quickly in stochastic anti-coordination games.

Keywords

Cite

@article{arxiv.1808.08416,
  title  = {Multiplayer bandits without observing collision information},
  author = {Gabor Lugosi and Abbas Mehrabian},
  journal= {arXiv preprint arXiv:1808.08416},
  year   = {2021}
}

Comments

To appear in Mathematics of Operations Research. 34 pages

R2 v1 2026-06-23T03:43:41.810Z