Timezone: »

 
Poster
Best of Both Worlds Model Selection
Aldo Pacchiano · Christoph Dann · Claudio Gentile

Wed Nov 30 09:00 AM -- 11:00 AM (PST) @ Hall J #802
in Poster Session 3 »

We study the problem of model selection in bandit scenarios in the presence of nested policy classes, with the goal of obtaining simultaneous adversarial and stochastic (``best of both worlds") high-probability regret guarantees. Our approach requires that each base learner comes with a candidate regret bound that may or may not hold, while our meta algorithm plays each base learner according to a schedule that keeps the base learner's candidate regret bounds balanced until they are detected to violate their guarantees. We develop careful mis-specification tests specifically designed to blend the above model selection criterion with the ability to leverage the (potentially benign) nature of the environment. We recover the model selection guarantees of the CORRAL algorithm for adversarial environments, but with the additional benefit of achieving high probability regret bounds. More importantly, our model selection results also hold simultaneously in stochastic environments under gap assumptions. These are the first theoretical results that achieve best-of-both world (stochastic and adversarial) guarantees while performing model selection in contextual bandit scenarios.

Keywords: Reinforcement Learning · bandits · stochastic · best of both worlds · adversarial · policy classes · model selection · linear bandits

Author Information

Aldo Pacchiano (Microsoft Research)
Christoph Dann (Google Research)
Claudio Gentile (Google Research)

More from the Same Authors