Regret Analysis of Stochastic and Nonstochastic Multi-armed Bandit Problems

TLDR

Multi‑armed bandit problems are foundational sequential decision problems balancing exploration and exploitation, with applications in ad placement, website optimization, and packet routing, and have been studied since the 1930s. This survey focuses on two extreme cases where regret analysis is particularly simple and elegant: i.i.d. payoffs and adversarial payoffs.

Abstract

Multi-armed bandit problems are the most basic examples of sequential decision problems with an exploration-exploitation trade-off. This is the balance between staying with the option that gave highest payoffs in the past and exploring new options that might give higher payoffs in the future. Although the study of bandit problems dates back to the Thirties, exploration-exploitation trade-offs arise in several modern applications, such as ad placement, website optimization, and packet routing. Mathematically, a multi-armed bandit is defined by the payoff process associated with each option. In this survey, we focus on two extreme cases in which the analysis of regret is particularly simple and elegant: i.i.d. payoffs and adversarial payoffs. Besides the basic setting of finitely many actions, we also analyze some of the most important variants and extensions, such as the contextual bandit model.