Research map / Bandits and online learning / Contextual bandits and Thompson sampling
BISTRO: An Efficient Relaxation-Based Method for Contextual Bandits
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| use BISTRO with random playout of future contexts and relaxation-based action selection | contextual bandits with i.i.d. contexts and arbitrary rewards · Hybrid i.i.d.-adversarial contextual bandits | expected regret: 2\sqrt{2d \cdot n \cdot \mathbb{E}\mathfrak{R}(\mathcal{F};x_{1:n})} | Helped |
| replace exact ERM values with δ-approximate values | BISTRO · Hybrid i.i.d.-adversarial contextual bandits | additional regret: O(n\delta) | Hurt |
| use a computationally easier superset relaxation for ERM | BISTRO · Contextual bandits with optimization-based ERM relaxations | regret bound: O(\sqrt{\mrm{gap}}) | Hurt |
| replace the factor 2 on random signs with a smaller value or zero (ablation) | BISTRO · Experiments (setting not specified) | performance | Helped |
| use BISTRO’s relaxation-based method with a full-information relaxation | adversarial contextual bandits · Adversarial contextual bandits | learnability | Helped |
Closest papers by meaning
- An Improved Relaxation for Oracle-Efficient Adversarial Contextual Bandits (NeurIPS 2023 · similarity 0.92)
- Efficient Algorithms for Adversarial Contextual Learning (ICML 2016 · similarity 0.92)
- Improved Regret Bounds for Oracle-Based Adversarial Contextual Bandits (NeurIPS 2016 · similarity 0.92)
- Beyond UCB: Optimal and Efficient Contextual Bandits with Regression Oracles (ICML 2020 · similarity 0.92)
- Active Context Selection Improves Simple Regret in Contextual Bandits (NeurIPS 2026 · similarity 0.90)
- Diffusion Models Meet Contextual Bandits (NeurIPS 2025 · similarity 0.89)
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.