Research map / Bandits and online learning / Contextual bandits and Thompson sampling
Thompson Sampling with Information Relaxation Penalties
Seungki Min, Costis Maglaras, Ciamac C. Moallemi
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 changed | Where | Measured | Effect |
|---|---|---|---|
| Replace TS with IRS.FH and then IRS.V-ZERO (progressively richer penalized information relaxation) | Thompson sampling · Gaussian MAB K=5 heteroscedastic noise | Bayesian regret | Improved |
| Use IRS policies instead of Bayes-UCB, IDS and OGI | IRS.V-ZERO · Gaussian MAB K=5 heteroscedastic noise | Bayesian regret | Improved |
| Use IRS.V-ZERO rather than IRS.FH (cost of better policy) | IRS policy · Gaussian MAB K=5, T=500 | simulation time per instance: 128 ms -> 7.4 sec | Hurt |
Closest papers by meaning
- Thompson Sampling with Less Exploration is Fast and Optimal (ICML 2023 · similarity 0.90)
- Variational Bayesian Optimistic Sampling (NeurIPS 2021 · similarity 0.90)
- Regret Bounds for Thompson Sampling in Episodic Restless Bandit Problems (NeurIPS 2019 · similarity 0.90)
- Batched Thompson Sampling (NeurIPS 2021 · similarity 0.89)
- Feel-Good Thompson Sampling for Contextual Bandits: a Markov Chain Monte Carlo Showdown (NeurIPS 2025 · similarity 0.88)
- The Choice of Noninformative Priors for Thompson Sampling in Multiparameter Bandit Models (AAAI 2024 · similarity 0.88)
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.
