AI Paper Atlas

Research map / Bandits and online learning / Online convex optimization and games

An $\alpha$-No-Regret Algorithm For Graphical Bilinear Bandits

Geovani Rizk, Igor Colin, Albert Thomas, Rida Laraki, Yann Chevaleyre

NeurIPS 2022 · Bandits and online learning › Online convex optimization and games · approach: Best-of-both-worlds bandit algorithms · 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
Select the optimistic node-arm pair using graph-weighted rewards for both cut edges and within-partition edges (ablation)OFUL for graphical bilinear bandits · Complete, random, and circle graphsalpha-regret guarantee (alpha): 0.505 + 0.495γ + ε vs. 0.5 + 0.5 γ; 0.547 + 0.453γ +ε vs. 0.5 + 0.5 γ; 0.99 + 0.01γ +ε vs. 0.5 + 0.5 γImproved
Use the improved graph-weighted pair selection (ablation)OFUL for graphical bilinear bandits · Star and matching graphsalpha-regret guarantee (alpha): 1 vs. 0.5 + 0.5 γImproved
Use the first OFUL variant, which selects pairs without weighting within-partition edges (ablation)Graphical bilinear bandits · Bipartite graphsalpha-regret guarantee (alpha): α=1 for the improved algorithm; no separate bipartite value for the first variantNo clear effect
Use the proposed algorithms instead of Explore-Then-ExploitSynthetic graphical bilinear bandit datasets · Synthetic datasetsfraction of global rewardImproved

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