AI Paper Atlas

Research map / Bandits and online learning / Multi-armed bandits

KL-UCB-Switch: Optimal Regret Bounds for Stochastic Bandits from Both a Distribution-Dependent and a Distribution-Free Viewpoints

Aurélien Garivier, Hédi Hadiji, Pierre Ménard, Gilles Stoltz

JMLR 2022 · Bandits and online learning › Multi-armed bandits · approach: Optimal stochastic bandit algorithms

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use switch exponent alpha=1/5 in KL-UCB-Switch instead of larger alpha (ablation)KL-UCB-Switch · distribution-dependent regret, three bandit instancescumulative regretNo clear effect
use switch exponent alpha=1/2 in KL-UCB-Switch (ablation)KL-UCB-Switch · distribution-dependent regret, three bandit instancescumulative regretMixed
use switch exponent alpha=8/9 in KL-UCB-Switch (ablation)KL-UCB-Switch · distribution-dependent regret, three bandit instancescumulative regretImproved
switch from KL-UCB to KL-UCB-Switch (any of three alphas)KL-UCB · distribution-free normalized regret R_T/sqrt(KT)normalized regretImproved
KL-UCB regret scaling with T and K versus KL-UCB-SwitchKL-UCB · distribution-free normalized regretscaling in T and KHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy