English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

We study best arm identification in a variant of the multi-armed banditproblem where the learner has limited precision in arm selection. The learnercan only sample arms via certain exploration bundles, which we refer to asboxes. In particular, at each sampling epoch, the learner selects a box, whichin turn causes an arm to get pulled as per a box-specific probabilitydistribution. The pulled arm and its instantaneous reward are revealed to thelearner, whose goal is to find the best arm by minimising the expected stoppingtime, subject to an upper bound on the error probability. We present anasymptotic lower bound on the expected stopping time, which holds as the errorprobability vanishes. We show that the optimal allocation suggested by thelower bound is, in general, non-unique and therefore challenging to track. Wepropose a modified tracking-based algorithm to handle non-unique optimalallocations, and demonstrate that it is asymptotically optimal. We also presentnon-asymptotic lower and upper bounds on the stopping time in the simplersetting when the arms accessible from one box do not overlap with those ofothers.

Best Arm Identification in Bandits with Limited Precision Sampling