AI Paper Atlas

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

Quantile Bandits for Best Arms Identification

Mengyan Zhang, Cheng Soon Ong

ICML 2021 · Bandits and online learning › Multi-armed bandits · approach: Best-arm identification in linear bandits · 3 citations (OpenAlex)

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
Q-SAR instead of Q-SR for multiple best arms (m=5)Q-SAR · Environment I and II, m=5probability of errorImproved
Q-SAR instead of Q-SR for single best arm (m=1)Q-SAR · Environment I and II, m=1probability of errorNo clear effect
Q-SAR instead of Q-Uniform and Q-BE baselinesQ-SAR · Environment I (0.5-quantile)probability of errorImproved
0.8-quantile based Q-SAR/Q-SR instead of mean-based SAR/SRQ-SAR · Environment IIprobability of errorImproved
Q-SAR instead of Q-SR at small sample sizeQ-SAR · Environment IIprobability of errorImproved
Use 0.8-quantile instead of mean as arm summary (larger min gap)Q-SAR · Environment IIprobability of error: 0.40 -> 1.21Improved
Q-SAR on vaccine simulation, m=3Q-SAR · Vaccine 0.5-quantileprobability of errorImproved
Q-SAR on vaccine simulation with outliers, m=1Q-SAR · Vaccine 0.5-quantileprobability of errorNo clear effect
Lower quantile level tauQ-SAR bound · problem complexity H_tauprobability of errorImproved

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