Research map / Bandits and online learning / Contextual bandits and Thompson sampling
No-Regret Algorithms for Heavy-Tailed Linear 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 |
|---|---|---|---|
| dynamic truncation of losses (alpha_t = t^{1/(2(1+eps))}) in the mean estimator | linear bandit confidence-region algorithm · synthetic heavy-tailed noise, T=10^6 | regret: about 25% lower regret | Helped |
| median-of-means estimator with batched updates (m = T^{2eps/(1+3eps)} batches) | linear bandit confidence-region algorithm · theoretical regret bound | regret exponent | Helped |
| median-of-means estimator in place of ridge regression in the regime eps→1 | linear bandit confidence-region algorithm · theoretical regret bound, all moments exist | regret exponent: eO(vT^{2/3}) vs eO(vT^{1/2}) for truncation | Hurt |
Closest papers by meaning
- Heavy-Tailed Linear Bandits: Huber Regression with One-Pass Update (ICML 2025 · similarity 0.93)
- Improved Regret Bounds for Linear Bandits with Heavy-Tailed Rewards (NeurIPS 2025 · similarity 0.92)
- Noise-Adaptive Confidence Sets for Linear Bandits and Application to Bayesian Optimization (ICML 2024 · similarity 0.91)
- Bayesian Regret Minimization in Offline Bandits (ICML 2024 · similarity 0.90)
- Taming Heavy-Tailed Losses in Adversarial Bandits and the Best-of-Both-Worlds Setting (NeurIPS 2024 · similarity 0.89)
- Improved Regret Analysis in Gaussian Process Bandits: Optimality for Noiseless Reward, RKHS norm, and Non-Stationary Variance (ICML 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.