AI Paper Atlas

Research map / Bandits and online learning / Contextual bandits and Thompson sampling

Exploiting Correlated Auxiliary Feedback in Parameterized Bandits

Arun Verma, Zhongxiang Dai, Yao Shu, Bryan Kian Hsiang Low

NeurIPS 2023 · Bandits and online learning › Contextual bandits and Thompson sampling · approach: Causal and structured bandits · 0 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
Replace observed rewards with hybrid rewards using known auxiliary feedbackOFUL, Lin-UCB, and NLin-UCB · Linear, linear contextual, and non-linear contextual banditsregretImproved
Estimate unknown auxiliary feedback functions using IS/MF samplingOFUL, Lin-UCB, and NLin-UCB variants · Linear, linear contextual, and non-linear contextual banditsregretImproved
Use a biased auxiliary feedback function estimateLin-UCB-BE · Linear contextual banditsregretMixed
Increase bias in the auxiliary feedback estimate (ablation)Lin-UCB-BE · Linear contextual banditsregretHurt
Increase the number of historical auxiliary feedback samples (ablation)Lin-UCB-EH · Linear contextual banditsregretImproved
Use an auxiliary feedback estimator trained with only a few historical samples (ablation)Lin-UCB-EH · Linear contextual banditsregretHurt
Increase reward and auxiliary-feedback correlation (ablation)Lin-UCB-AF · Linear contextual banditsregretImproved

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