Skip to yearly menu bar Skip to main content


Poster

Improving Regret Bounds for Combinatorial Semi-Bandits with Probabilistically Triggered Arms and Its Applications

Qinshi Wang · Wei Chen

Pacific Ballroom #63

Keywords: [ Bandit Algorithms ] [ Online Learning ] [ Reinforcement Learning ]


Abstract:

We study combinatorial multi-armed bandit with probabilistically triggered arms (CMAB-T) and semi-bandit feedback. We resolve a serious issue in the prior CMAB-T studies where the regret bounds contain a possibly exponentially large factor of 1/p, where p is the minimum positive probability that an arm is triggered by any action. We address this issue by introducing a triggering probability modulated (TPM) bounded smoothness condition into the influence maximization bandit and combinatorial cascading bandit satisfy this TPM condition. As a result, we completely remove the factor of 1/p* from the regret bounds, achieving significantly better regret bounds for influence maximization and cascading bandits than before. Finally, we provide lower bound results showing that the factor 1/p* is unavoidable for general CMAB-T problems, suggesting that the TPM condition is crucial in removing this factor.

Live content is unavailable. Log in and register to view live content