Off-policy Bandits with Deficient Support
This addresses a critical gap for applications like voice assistants and recommendation systems where large action spaces often lead to support-deficient data, offering incremental improvements to existing methods.
The paper tackles the problem of off-policy learning in contextual bandits when data lacks full support, showing that existing methods can fail catastrophically, and it analyzes three approaches—restricting action space, reward extrapolation, and restricting policy space—with empirical evaluations to provide practical recommendations.
Learning effective contextual-bandit policies from past actions of a deployed system is highly desirable in many settings (e.g. voice assistants, recommendation, search), since it enables the reuse of large amounts of log data. State-of-the-art methods for such off-policy learning, however, are based on inverse propensity score (IPS) weighting. A key theoretical requirement of IPS weighting is that the policy that logged the data has "full support", which typically translates into requiring non-zero probability for any action in any context. Unfortunately, many real-world systems produce support deficient data, especially when the action space is large, and we show how existing methods can fail catastrophically. To overcome this gap between theory and applications, we identify three approaches that provide various guarantees for IPS-based learning despite the inherent limitations of support-deficient data: restricting the action space, reward extrapolation, and restricting the policy space. We systematically analyze the statistical and computational properties of these three approaches, and we empirically evaluate their effectiveness. In addition to providing the first systematic analysis of support-deficiency in contextual-bandit learning, we conclude with recommendations that provide practical guidance.