Alekh Agarwal, Daniel Hsu, Satyen Kale, John Langford, Lihong Li, and Robert E. Schapire
Proceedings of 31st International Conference on Machine Learning (ICML), 2014

We present a new algorithm for the contextual bandit learning problem, where the learner repeatedly takes one of \(K\) actions in response to the observed context, and observes the reward only for that chosen action. Our method assumes access to an oracle for solving fully supervised cost-sensitive classification problems and achieves the statistically optimal regret guarantee with only \(\tilde{O}(\sqrt{KT})\) oracle calls across all \(T\) rounds. By doing so, we obtain the most practical contextual bandit learning algorithm amongst approaches that work for general policy classes. We further conduct a proof-of-concept experiment which demonstrates the excellent computational and prediction performance of (an online variant of) our algorithm relative to several baselines.