Communication-Efficient Collaborative Regret Minimization in Multi-Armed Bandits

TL;DR

This paper explores the balance between communication frequency and regret minimization in multi-agent multi-armed bandit problems, providing new tradeoff insights for collaborative learning efficiency.

Contribution

It introduces the first tradeoff analysis between communication rounds and regret in collaborative multi-armed bandit settings.

Findings

Established fundamental tradeoffs between communication and regret

Provided theoretical bounds for collaborative learning efficiency

Analyzed the impact of communication frequency on regret minimization

Abstract

In this paper, we study the collaborative learning model, which concerns the tradeoff between parallelism and communication overhead in multi-agent multi-armed bandits. For regret minimization in multi-armed bandits, we present the first set of tradeoffs between the number of rounds of communication among the agents and the regret of the collaborative learning process.