Chaterjee, A and Ghalme, G and Jain, S and Vaish, R and Narahari, Y (2017) Analysis of Thompson sampling for stochastic sleeping bandits. In: 33rd Conference on Uncertainty in Artificial Intelligence, UAI 2017, 11 - 15 August 2017, Sydney.
PDF
UAI 2017_2017.pdf - Published Version Restricted to Registered users only Download (523kB) | Request a copy |
Abstract
We study a variant of the stochastic multiarmed bandit problem where the set of available arms varies arbitrarily with time (also known as the sleeping bandit problem). We focus on the Thompson Sampling algorithm and consider a regret notion defined with respect to the best available arm. Our main result is an O(log T) regret bound for Thompson Sampling, which generalizes a similar bound known for this algorithm from the classical bandit setting. Our bound also matches (up to constants) the best-known lower bound for the sleeping bandit problem. We show via simulations that Thompson Sampling outperforms the UCB-style AUER algorithm for the sleeping bandit problem.
Item Type: | Conference Paper |
---|---|
Publication: | Uncertainty in Artificial Intelligence - Proceedings of the 33rd Conference, UAI 2017 |
Publisher: | AUAI Press Corvallis |
Additional Information: | The copyright for this article belongs to the AUAI Press Corvallis. |
Keywords: | Artificial intelligence; Probability; Stochastic systems, Bandit problems; Lower bounds; Multi-armed bandit problem; Regret bounds; Thompson samplings, Sleep research |
Department/Centre: | Division of Electrical Sciences > Computer Science & Automation |
Date Deposited: | 22 Jul 2022 11:18 |
Last Modified: | 22 Jul 2022 11:18 |
URI: | https://eprints.iisc.ac.in/id/eprint/74737 |
Actions (login required)
View Item |