Efficient Contextual Bandits with Continuous Actions

We create a computationally tractable learning algorithm for contextual bandits with continuous actions having unknown structure. The new reduction-style algorithm composes with most supervised learning representations. We prove that this algorithm works in a general sense and verify the new functionality with large-scale experiments.

Authors

• Maryam Majzoubi
• Chicheng Zhang
• Rajan Chari
• Akshay Krishnamurthy
• John Langford
• Aleksandrs Slivkins Microsoft

Keywords

No keywords are indexed for this paper.