AI Paper Atlas

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

Dueling Bandits with Adversarial Sleeping

Aadirupa Saha, Pierre Gaillard

NeurIPS 2021 · Bandits and online learning › Multi-armed bandits · approach: Non-stationary and structured-reward 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
Use pairwise confidence bounds and availability-aware candidate selectionRUC-BP · DB-SPAAfixed-confidence regret: R_T \leq 2 \sum_{i=1}^{K-1} \sum_{j=i+1}^K M_{ij} \log \big(2 C(K,\delta) M_{ij}\big)Improved
Use empirical-divergence filtering and select uniformly from near-best itemsR-MED · DB-SPAAexpected regret: O\bigg( \min \bigg\{ \sum_{j = 2}^{K} \frac{K \log T}{\Delta(j-1,j)_+}, \ K T^{2/3} \bigg\} \bigg)Improved
Use R-MED rather than RUC-BPThe proposed algorithms · Empirical evaluations across preference and availability instancescumulative regretMixed
Use low-correlation rather than high-correlation availability sequencesRUC-BP and R-MED on the Hard preference instance · Availability sequencescumulative regretImproved

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