AI Paper Atlas

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

A KL-LUCB algorithm for Large-Scale Crowdsourcing

Ervin Tanczos, Robert Nowak, Bob Mankoff

NeurIPS 2017 · Bandits and online learning › Multi-armed bandits · approach: Best-arm identification in linear bandits

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 KL-based confidence bounds instead of sub-Gaussian boundslil-KLUCB · Synthetic Bernoulli rewards, α=1samples to find the best arm: about 20 % fewer samplesImproved
Use KL-based confidence bounds instead of sub-Gaussian boundslil-KLUCB · Synthetic Bernoulli rewards, α=1/2samples to find the best arm: 30 % fewer samplesImproved
Use KL-based confidence bounds instead of sub-Gaussian boundslil-KLUCB · New Yorker Caption Contest data, best-arm mean about 0.5empirical probability that the best arm is among the top 5No clear effect
Use KL-based confidence bounds instead of sub-Gaussian boundslil-KLUCB · New Yorker Caption Contest data, high best-arm meanempirical probability that the best arm is among the top 5Improved
Use KL-based rather than sub-Gaussian confidence boundsBest-arm identification with gaps Δᵢ=(i/n)^α · Theoretical sample-complexity scaling, α=1sample complexity: n log n vs n²Improved
Use KL-based rather than sub-Gaussian confidence boundsBest-arm identification with gaps Δᵢ=(i/n)^α · Theoretical sample-complexity scaling, α∈(1/2,1)sample complexity: n vs n^(2α)Improved
Use lil-KLUCB’s confidence-bound analysis rather than KL-LUCB’sBest-arm identification with bounded rewards · Fixed-confidence best-arm identificationsample-complexity dependence on arm count and Chernoff informationImproved

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