Skip to yearly menu bar Skip to main content


Poster

Sparsity-Agnostic Linear Bandits with Adaptive Adversaries

Tianyuan Jin · Kyoungseok Jang · Nicolò Cesa-Bianchi

[ ] [ Project Page ]
Wed 11 Dec 4:30 p.m. PST — 7:30 p.m. PST

Abstract: We study stochastic linear bandits where, in each round, the learner receives a set of actions (i.e., feature vectors), from which it chooses an element and obtains a stochastic reward. The expected reward is a fixed but unknown linear function of the chosen action. We study \emph{sparse} regret bounds, that depend on the number $S$ of non-zero coefficients in the linear reward function. Previous works focused on the case where $S$ is known, or the action sets satisfy additional assumptions. In this work, we obtain the first sparse regret bounds that hold when $S$ is unknown and the action sets are adversarially generated. Our techniques combine online to confidence set conversions with a novel randomized model selection approach over a hierarchy of nested confidence sets. When $S$ is known, our analysis recovers state-of-the-art bounds for adversarial action sets. We also show that a variant of our approach, using Exp3 to dynamically select the confidence sets, can be used to improve the empirical performance of stochastic linear bandits while enjoying a regret bound with optimal dependence on the time horizon.

Live content is unavailable. Log in and register to view live content