Thompson Sampling

Ask about this topic

Overview

Thompson sampling maintains a probability model of uncertain rewards, samples a plausible model, and chooses the action that looks best under that sample. It is studied in multi-armed and contextual bandits, linear and combinatorial settings, and Bayesian optimization with Gaussian processes; methods range from conjugate posteriors to nonparametric models and approximate sampling algorithms. Regret analyses give guarantees for many settings, while practical work adapts the method to changing rewards, structured actions, privacy, and limited computation. Open tensions include sensitivity to priors and model misspecification, the cost of accurate posterior sampling, and how to retain strong guarantees under non-stationarity or complex feedback.

Latest papers

CardsList

53 papers10 papers in the last four weeks, up 100% on the four weeks before. 0.1% of all new papers.

  1. Exploratory creative pre-ranking: Hierarchical empirical-Bayes Thompson sampling for real-time creative optimization

    Oct 26, 2026Jiale Han, Ajith Moparthi, Dan Xie +1Learning to RankThompson Sampling

    Department of Statistics and Data Science, UCLA · Roblox · Departments of Statistics and Data Science and of Biostatistics, UCLA

  2. Amortized Thompson Sampling and Efficient Information-Directed Sampling for Simulator-Based Design

    Oct 23, 2026Michal Kobiela, Michael U. GutmannThompson SamplingBayesian Experimental Design

    School of Informatics The University of Edinburgh Edinburgh, UK

  3. Beyond Expected Values: Risk-Sensitive Planning with Distributional Monte-Carlo Tree Search

    Oct 16, 2026Quynh Trinh, Trung Nguyen, Nam Nguyen +1Risk-Sensitive PlanningMonte Carlo Tree Search

    Hanoi University of Science and Technology, Hanoi, Vietnam · Center for AI Research, VinUniversity, Hanoi, Vietnam

  4. The Tilted Sampling Problem: Optimism, Posterior Sampling, and Logarithmic Regret

    Oct 15, 2026Mohammad Pedramfar, Siamak RavanbakhshPosterior SamplingThompson Sampling

    Mila – Québec AI Institute, McGill University

  5. Convergence Rates of Gaussian Process Thompson Sampling for Efficient Global Optimization

    Oct 14, 2026Haowei Wang, Yilin ZhengThompson SamplingGlobal Optimization

    National University of Singapore Singapore

  6. High-Probability Regret Bounds for Feel-Good Thompson Sampling

    Oct 14, 2026Xuheng Li, Qingyue Zhao, Quanquan GuThompson SamplingContextual Bandits

    Department of Computer Science, University of California, Los Angeles, CA 90095, USA

  7. Expected Sample Complexity in Multi-Armed Bandits

    Oct 7, 2026Nadav Sukenik, Nadav MerlisMulti-Armed BanditsRegret Minimization

    Technion – Institute of Technology

  8. Structure, Not Belief: Correlated Thompson Sampling from LLM-Derived Covariance in Combinatorial Semi-Bandits

    Oct 5, 2026Vikram Kakaria, Anish Kataria, Anany KotawalaThompson SamplingMulti-Armed Bandits

    Princeton University

  9. Ranking Bandits for Carousel Interfaces with Observable Browsing Depth

    Oct 4, 2026Takuma Yasuda, Atsuyoshi Nakamura, Dai Hai NguyenThompson SamplingMulti-Armed Bandits

    Graduate School of Information Science and Technology, Hokkaido University, Sapporo, Japan

  10. When May a Bandit Leave Its Anchor? E-Process-Authorized Thompson Sampling under Non-stationarity

    Oct 2, 2026Mayand Gulati, Kerong Wang, WeiChen AuThompson SamplingNon-Stationary Bandits

    UC Santa Barbara · Purdue University

  11. Risk-Aware Adaptive Evaluation: Finding High-Impact Failures Under Limited Budgets

    Sep 30, 2026Priyanath Maji, Spandan Ghose ChowdhuryThompson SamplingAI Agent Evaluation

    Georgia Institute of Technology Atlanta, GA 30332

  12. A Unified Optimism-Agnostic Framework for Linear Bandits over Spherical Action Sets

    Sep 26, 2026Arda Güçlü, Subhonmesh Bose, John R. BirgeMulti-Armed BanditsUCB Algorithms

    University of Chicago Booth School of Business, Chicago, IL 60637

  13. Exact Bayes Regret and Asymptotic Optimality in High-Dimensional Gaussian Bandits

    Sep 23, 2026Prakhar Singhvi, Yi Zou, Abhishek BhattacharjeeThompson SamplingMulti-Armed Bandits

    Abstract Math Institute

  14. Odds-Ratio Thompson Sampling: A Specification and Design Guide for Contrast-Based Multi-Armed Bandits

    Sep 17, 2026Sulgi KimThompson SamplingMulti-Armed Bandits

    Independent researcher

  15. Thompson Sampling for Non-Monotone Convex Ridge Bandits: Monotonicity Is Not Needed for Polynomial Regret

    Sep 10, 2026Xuan LiThompson SamplingBandit Convex Optimization

    University of New South Wales, Sydney, Australia

  16. Posterior Tempering Explains Variance Inflation in Linear and Generalized Linear Thompson Sampling

    Sep 2, 2026Prateek Jaiswal, Debdeep Pati, Anirban Bhattacharya +1Thompson SamplingMulti-Armed Bandits

    Daniels School of Business, Purdue University. · Department of Statistics, University of Wisconsin-Madison. · Department of Statistics, Texas A&M University.

  17. Improved Regret Analysis for Parallel Gaussian Process Bandit Optimization

    Aug 17, 2026Shion Takeno, Shogo IwazakiThompson SamplingMulti-Armed Bandits

    Nagoya University · MI-6 Ltd.

  18. LLM-Derived Priors for Thompson Sampling in Cold-Start Comment Recommendation

    Aug 4, 2026Eugene Lee, Oseong Choi, Byungsoo Kang +1Thompson SamplingCold-Start Recommendation

    NAVER WEBTOON South Korea

  19. On-Policy and Off-Policy Learning for Large Action Spaces

    Jul 30, 2026Imad AoualiThompson SamplingMulti-Armed Bandits

    Institut Polytechnique de Paris · Ecole nationale de la statistique et de l’administration économique · Ecole doctorale n◦574 Ecole Doctorale de Mathématique Hadamard (EDMH)

  20. Periodic Bootstrap Thompson Sampling For Periodically Non-Stationary Bandit Problems

    Jul 18, 2026Boning ShaoThompson SamplingNon-Stationary Bandits

    Faculty of Science, University of Hong Kong, Hong Kong, China

  21. Information-Directed Sampling for Causal Bandits

    Jul 17, 2026Muhammad Qasim Elahi, Murat Kocaoglu, Mahsa GhasemiThompson SamplingMulti-Armed Bandits

    School of Electrical and Computer Engineering, Purdue University · School of Computer Science, Johns Hopkins University

  22. Thompson Sampling Is 2-Competitive for Mistakes

    Jul 14, 2026Mark Sellke, Gregory ValiantThompson SamplingMulti-Armed Bandits

  23. Link Adaptation Using Joint-Thompson Sampling

    Jul 13, 2026Vignatha Vinjam, Manjunath Kolavennu, Myna Vajha +1Thompson SamplingMulti-Armed Bandits

    Department of EE, IIT Hyderabad · Department of AI, IIT Hyderabad

  24. Randomized Exploration for Linear Bandits via Absolute Perturbations

    Jun 26, 2026Toshinori Kitamura, Shuai Liu, Csaba SzepesváriThompson SamplingMulti-Armed Bandits

    Department of Computing Science, University of Alberta, Canada

  25. Flow-Corrected Thompson Sampling for Non-Stationary Contextual Bandits

    Jun 22, 2026AmirHossein Naghdi, Ali BaheriThompson SamplingNon-Stationary Bandits

    Sharif University of Technology, Iran · Rochester Institute of Technology, USA

  26. AdaPrivate-TS: Private Thompson Sampling for Contextual Bandits with Privacy Amplification

    Jun 19, 2026Mohammadreza Riyazat, Eranga UkwattaThompson SamplingDifferential Privacy

    School of Engineering, University of Guelph, Guelph, ON, Canada.

  27. Bayesian Anytime Pareto Set Identification for Multi-Objective Multi-Armed Bandits

    Jun 17, 2026Lennert Saerens, Bram Silue, Eleni Litsa +2Thompson SamplingMulti-Armed Bandits

    Artificial Intelligence Lab, Departement of Computer Science, Vrije Universiteit Brussel, Brussels, Belgium · imec, Leuven, Belgium · Data Science Institute, Interuniversity Institute of Biostatistics and Statistical Bioinformatics, UHasselt, Hasselt, Belgium

  28. Asymptotic Optimality of Thompson Sampling for Risk-Averse Bandits with Sub-Gaussian Rewards

    Jun 8, 2026Joel Q. L. ChangThompson SamplingMulti-Armed Bandits

  29. MINTS: Minimalist Thompson Sampling

    Jun 1, 2026Kaizheng WangThompson SamplingMulti-Armed Bandits

    Department of IEOR and Data Science Institute, Columbia University.

  30. Variance-sensitive Thompson sampling for generalised linear bandits, revisited

    May 29, 2026Tom Perneczky, Marc Abeille, David JanzThompson SamplingMulti-Armed Bandits

    University of Oxford · Criteo AI Lab