Collaborative Regret Minimization in Multi-Armed Bandits

01/26/2023

∙

by   Nikolai Karpov, et al.

∙

0

∙

share

In this paper, we study the collaborative learning model, which concerns the tradeoff between parallelism and communication overhead in multi-agent reinforcement learning. For a fundamental problem in bandit theory, regret minimization in multi-armed bandits, we present the first and almost tight tradeoffs between the number of rounds of communication between the agents and the regret of the collaborative learning process.