Cooperative Multi-Agent Bandits with Heavy Tails

Abstract

We study the heavy-tailed stochastic bandit problem in the cooperative multi-agent setting, where a group of agents interact with a common bandit problem, while communicating on a network with delays. Existing algorithms for the stochastic bandit in this setting utilize confidence intervals arising from an averaging-based communication protocol known as running consensus, that does not lend itself to robust estimation for heavy-tailed settings. We propose MP-UCB, a decentralized multi-agent algorithm for the cooperative stochastic bandit that incorporates robust estimation with a message-passing protocol. We prove optimal regret bounds for MP-UCB for several problem settings, and also demonstrate its superiority to existing methods. Furthermore, we establish the first lower bounds for the cooperative bandit problem, in addition to providing efficient algorithms for robust bandit estimation of location.

Cite

Text

Dubey and Pentland. "Cooperative Multi-Agent Bandits with Heavy Tails." International Conference on Machine Learning, 2020.

Markdown

[Dubey and Pentland. "Cooperative Multi-Agent Bandits with Heavy Tails." International Conference on Machine Learning, 2020.](https://mlanthology.org/icml/2020/dubey2020icml-cooperative/)

BibTeX

@inproceedings{dubey2020icml-cooperative,
  title     = {{Cooperative Multi-Agent Bandits with Heavy Tails}},
  author    = {Dubey, Abhimanyu and Pentland, Alex ‘Sandy’},
  booktitle = {International Conference on Machine Learning},
  year      = {2020},
  pages     = {2730-2739},
  volume    = {119},
  url       = {https://mlanthology.org/icml/2020/dubey2020icml-cooperative/}
}