Analysis of thompson sampling for combinatorial multi-armed bandit with probabilistically triggered arms
Proceedings of the 22nd International Conference on Artificial Intelligence and Statistics, AISTATS 2019
Item Usage Stats
We analyze the regret of combinatorial Thompson sampling (CTS) for the combinatorial multi-armed bandit with probabilistically triggered arms under the semi-bandit feedback setting. We assume that the learner has access to an exact optimization oracle but does not know the expected base arm outcomes beforehand. When the expected reward function is Lipschitz continuous in the expected base arm outcomes, we derive O( Pm i=1 log T /(pii)) regret bound for CTS, where m denotes the number of base arms, pi denotes the minimum non-zero triggering probability of base arm i and i denotes the minimum suboptimality gap of base arm i. We also compare CTS with combinatorial upper confidence bound (CUCB) via numerical experiments on a cascading bandit problem.