A Diffusion Analysis of Policy Gradient for Stochastic Bandits

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autore principale: Lattimore, Tor
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866918381881393152
author Lattimore, Tor
author_facet Lattimore, Tor
contents We study a continuous-time diffusion approximation of policy gradient for $k$-armed stochastic bandits. We prove that with a learning rate $η= O(Δ^2/\log(n))$ the regret is $O(k \log(k) \log(n) / η)$ where $n$ is the horizon and $Δ$ the minimum gap. Moreover, we construct an instance with only logarithmically many arms for which the regret is linear unless $η= O(Δ^2)$.
format Preprint
id arxiv_https___arxiv_org_abs_2603_10219
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle A Diffusion Analysis of Policy Gradient for Stochastic Bandits
Lattimore, Tor
Machine Learning
Artificial Intelligence
Statistics Theory
We study a continuous-time diffusion approximation of policy gradient for $k$-armed stochastic bandits. We prove that with a learning rate $η= O(Δ^2/\log(n))$ the regret is $O(k \log(k) \log(n) / η)$ where $n$ is the horizon and $Δ$ the minimum gap. Moreover, we construct an instance with only logarithmically many arms for which the regret is linear unless $η= O(Δ^2)$.
title A Diffusion Analysis of Policy Gradient for Stochastic Bandits
topic Machine Learning
Artificial Intelligence
Statistics Theory
url https://arxiv.org/abs/2603.10219