Thompson Sampling with a Mixture Prior

Joey Hong · Branislav Kveton · Manzil Zaheer · Mohammad Ghavamzadeh · Craig Boutilier

[ Abstract ]
Mon 28 Mar 10:15 a.m. PDT — 11:45 a.m. PDT


We study Thompson sampling (TS) in online decision making, where the uncertain environment is sampled from a mixture distribution. This is relevant in multi-task learning, where a learning agent faces different classes of problems. We incorporate this structure in a natural way by initializing TS with a mixture prior, and call the resulting algorithm MixTS. To analyze MixTS, we develop a novel and general proof technique for analyzing the concentration of mixture distributions. We use it to derive Bayes regret bounds for MixTS in both linear bandits and finite-horizon reinforcement learning (RL). Our regret bounds reflect the structure of the mixture prior, and depend on the number of mixture components and their width. We demonstrate the empirical effectiveness of MixTS in synthetic and real-world experiments.

Chat is not available.