Mathematics – Statistics Theory
Scientific paper
2008-05-22
Mathematics
Statistics Theory
24 pages
Scientific paper
Multi-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.
Garivier Aurelien
Moulines Eric
No associations
LandOfFree
On Upper-Confidence Bound Policies for Non-Stationary Bandit Problems does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with On Upper-Confidence Bound Policies for Non-Stationary Bandit Problems, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and On Upper-Confidence Bound Policies for Non-Stationary Bandit Problems will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-510690