2013 COLT COLT 2013

Information Complexity in Bandit Subset Selection

Abstract

We consider the problem of efficiently exploring the arms of a stochastic bandit to identify the best subset. Under the PAC and the fixed-budget formulations, we derive improved bounds by using KL-divergence-based confidence intervals. Whereas the application of a similar idea in the regret setting has yielded bounds in terms of the KL-divergence between the arms, our bounds in the pure-exploration setting involve the Chernoff information between the arms. In addition to introducing this novel quantity to the bandits literature, we contribute a comparison between the “racing” and “smart sampling” strategies for pure-exploration problems, finding strong evidence in favor of the latter.

🌉 Interdisciplinary Bridge — Machine Learning and Mathematics & Optimization
🧭 Keyword Pioneer — pure exploration
🐝 Cross-Pollinator — Artificial Intelligence, Knowledge & Reasoning, Machine Learning, Mathematics & Optimization, Natural Language Processing, Reinforcement Learning, Security & Privacy
📈 Trend Setter — Evaluation
🐣 Hot Topic Early Bird — pac learning