Timezone: »

 
Spotlight
Contextual Bandits with Large Action Spaces: Made Practical
Yinglun Zhu · Dylan Foster · John Langford · Paul Mineiro

Wed Jul 20 08:20 AM -- 08:25 AM (PDT) @ Hall F
in T: Learning Theory/Domain Adaptation »

A central problem in sequential decision making is to develop algorithms that are practical and computationally efficient, yet support the use of flexible, general-purpose models. Focusing on the contextual bandit problem, recent progress provides provably efficient algorithms with strong empirical performance when the number of possible alternatives (``actions'') is small, but guarantees for decision making in large, continuous action spaces have remained elusive, leading to a significant gap between theory and practice. We present the first efficient, general-purpose algorithm for contextual bandits with continuous, linearly structured action spaces. Our algorithm makes use of computational oracles for (i) supervised learning, and (ii) optimization over the action space, and achieves sample complexity, runtime, and memory independent of the size of the action space. In addition, it is simple and practical. We perform a large-scale empirical evaluation, and show that our approach typically enjoys superior performance and efficiency compared to standard baselines.

Keywords: T: Online Learning and Bandits

Author Information

Yinglun Zhu (University of Wisconsin-Madison)
Dylan Foster (Microsoft Research)
John Langford (Microsoft Research)
Paul Mineiro (Microsoft)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors