We consider the problem of best Markovian arm identification, where we sequentially collect samples from K Markov chains with our goal being to identify the one with the largest stationary mean with some fixed level of confidence. In Theorem 4 we derive an instance specific non-asymptotic lower bound for the sample complexity, which in the high confidence regime (Corollary 5) generalizes the asymptotic lower bound of Garivier and Kaufmann (2016) which deals with the special case where the K stochastic processes are i.i.d. processes.
Title
A Lower Bound for Identifying the Best Markovian Arm with Fixed Confidence
Published
2019-05-14
Full Collection Name
Electrical Engineering & Computer Sciences Technical Reports
Other Identifiers
EECS-2019-33
Type
Text
Extent
15 p
Archive
The Engineering Library
Usage Statement
Researchers may make free and open use of the UC Berkeley Library’s digitized public domain materials. However, some materials in our online collections may be protected by U.S. copyright law (Title 17, U.S.C.). Use or reproduction of materials protected by copyright beyond that allowed by fair use (Title 17, U.S.C. § 107) requires permission from the copyright owners. The use or reproduction of some materials may also be restricted by terms of University of California gift or purchase agreements, privacy and publicity rights, or trademark law. Responsibility for determining rights status and permissibility of any use or reproduction rests exclusively with the researcher. To learn more or make inquiries, please see our permissions policies (https://www.lib.berkeley.edu/about/permissions-policies).
