On Upper-Confidence Bound Policies for Non-Stationary Bandit Problems

dc.creatorGarivier, Aurélien
dc.creatorMoulines, Eric
dc.date2008-05-22
dc.date.accessioned2026-07-07T12:19:10Z
dc.date.available2026-07-07T12:19:10Z
dc.descriptionMulti-armed bandit problems are considered as a paradigm of the trade-off between exploring the environment to find profitable actions and exploiting what is already known. In the stationary case, the distributions of the rewards do not change in time, Upper-Confidence Bound (UCB) policies have been shown to be rate optimal. A challenging variant of the MABP is the non-stationary bandit problem where the gambler must decide which arm to play while facing the possibility of a changing environment. In this paper, we consider the situation where the distributions of rewards remain constant over epochs and change at unknown time instants. We analyze two algorithms: the discounted UCB and the sliding-window UCB. We establish for these two algorithms an upper-bound for the expected regret by upper-bounding the expectation of the number of times a suboptimal arm is played. For that purpose, we derive a Hoeffding type inequality for self normalized deviations with a random number of summands. We establish a lower-bound for the regret in presence of abrupt changes in the arms reward distributions. We show that the discounted UCB and the sliding-window UCB both match the lower-bound up to a logarithmic factor.
dc.description24 pages
dc.identifierhttps://arxiv.org/abs/0805.3415
dc.identifierhttp://arxiv.org/abs/0805.3415
dc.identifier.urihttp://salesiana.dossiersoluciones.com/handle/123456789/212661
dc.subjectStatistics Theory
dc.titleOn Upper-Confidence Bound Policies for Non-Stationary Bandit Problems
dc.typetext

Files

Collections