Near-Optimal Regret Bounds for Thompson Sampling

• Published in: Journal of the ACM Vol. 64; no. 5; pp. 1 - 24
• Main Authors: Agrawal, Shipra, Goyal, Navin
• Format: Journal Article
• Language: English
• Published: New York Association for Computing Machinery 01.10.2017
• Subjects: Bayesian analysis; Empirical analysis; Heuristic; Lower bounds; Martingales; Optimization; Probability distribution functions; Randomized algorithms; Sampling; Studies
• ISSN: 0004-5411; 1557-735X
• DOI: 10.1145/3088510

Thompson Sampling (TS) is one of the oldest heuristics for multiarmed bandit problems. It is a randomized algorithm based on Bayesian ideas and has recently generated significant interest after several studies demonstrated that it has favorable empirical performance compared to the state-of-the-art...