A Tight Lower Bound for Non-stochastic Multi-armed Bandits with Expert Advice

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Chase, Zachary, Ito, Shinji, Mehalel, Idan
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915589732171776
author Chase, Zachary
Ito, Shinji
Mehalel, Idan
author_facet Chase, Zachary
Ito, Shinji
Mehalel, Idan
contents We determine the minimax optimal expected regret in the classic non-stochastic multi-armed bandit with expert advice problem, by proving a lower bound that matches the upper bound of Kale (2014). The two bounds determine the minimax optimal expected regret to be $Θ\left( \sqrt{T K \log (N/K) } \right)$, where $K$ is the number of arms, $N$ is the number of experts, and $T$ is the time horizon.
format Preprint
id arxiv_https___arxiv_org_abs_2511_00257
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Tight Lower Bound for Non-stochastic Multi-armed Bandits with Expert Advice
Chase, Zachary
Ito, Shinji
Mehalel, Idan
Machine Learning
We determine the minimax optimal expected regret in the classic non-stochastic multi-armed bandit with expert advice problem, by proving a lower bound that matches the upper bound of Kale (2014). The two bounds determine the minimax optimal expected regret to be $Θ\left( \sqrt{T K \log (N/K) } \right)$, where $K$ is the number of arms, $N$ is the number of experts, and $T$ is the time horizon.
title A Tight Lower Bound for Non-stochastic Multi-armed Bandits with Expert Advice
topic Machine Learning
url https://arxiv.org/abs/2511.00257