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

In this paper we propose a general methodology to derive regret bounds forrandomized multi-armed bandit algorithms. It consists in checking a set ofsufficient conditions on the sampling probability of each arm and on the familyof distributions to prove a logarithmic regret. As a direct application werevisit two famous bandit algorithms, Minimum Empirical Divergence (MED) andThompson Sampling (TS), under various models for the distributions includingsingle parameter exponential families, Gaussian distributions, boundeddistributions, or distributions satisfying some conditions on their moments. Inparticular, we prove that MED is asymptotically optimal for all these models,but also provide a simple regret analysis of some TS algorithms for which theoptimality is already known. We then further illustrate the interest of ourapproach, by analyzing a new Non-Parametric TS algorithm (h-NPTS), adapted tosome families of unbounded reward distributions with a bounded h-moment. Thismodel can for instance capture some non-parametric families of distributionswhose variance is upper bounded by a known constant.

A General Recipe for the Analysis of Randomized Multi-Armed Bandit Algorithms