Work

Fixed-Confidence Guarantees for Bayesian Best-Arm Identification

Year: 2019

Type: conference-paper

Abstract: We investigate and provide new insights on the sampling rule called Top-Two Thompson Sampling (TTTS). In particular, we justify its use for fixed-confidence best-arm identification. We further propose a variant of TTTS called Top-Two Transportation Cost (T3C), which disposes of the computational burden of TTTS. As our main contribution, we provide the first sample complexity analysis of TTTS and T3C when coupled with a very natural Bayesian stopping rule, for bandits with Gaussian rewards, solving one of the open questions raised by Russo (2016). We also provide new posterior convergence results for TTTS under two models that are commonly used in practice: bandits with Gaussian and Bernoulli rewards and conjugate priors.

Source: Leiden Repository (Leiden University)

DOI: 10.48550/arxiv.1910.10945

Authors: Shang, Xuedong, de Heide, Rianne, Kaufmann, Emilie, Ménard, Pierre, Valko, Michal

Institutions: Centre de Recherche en Informatique, Signal et Automatique de Lille, Leiden University, Centrum Wiskunde & Informatica, Centre National de la Recherche Scientifique, Google DeepMind (United Kingdom)

Language: English

Topic: Advanced Bandit Algorithms Research

Subfield: Management Science and Operations Research

Field: Decision Sciences

Domain: Social Sciences

Open Access status: green

Funders: Agence Nationale de la Recherche, CHIST-ERA

Awards: G3010527862, BANDITS AGAINST NON-STATIONARITY AND STRUCTURE, Beyond Online Learning for better Decision making