Efficient Contextual Bandits with Continuous Actions

Abstract

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

arXiv commentTo appear at NeurIPS 2020

References in corpus (14)

• A Contextual-Bandit Approach to Personalized News Article Recommendation
• Doubly Robust Policy Evaluation and Learning
• Taming the Monster: A Fast and Simple Algorithm for Contextual Bandits
• Contextual Bandits with Similarity Information
• Counterfactual Risk Minimization: Learning from Logged Bandit Feedback
• Efficient Optimal Learning for Contextual Bandits
• Making Contextual Decisions with Low Technical Debt
• BISTRO: An Efficient Relaxation-Based Method for Contextual Bandits
• Policy Evaluation and Optimization with Continuous Treatments
• Black-box optimization of noisy functions with unknown smoothness
• A Contextual Bandit Bake-off
• Efficient Algorithms for Adversarial Contextual Learning
• On the Optimality of Trees Generated by ID3
• Top-down induction of decision trees: rigorous guarantees and inherent limitations

Cited by in corpus (2)

• Probabilistic design of optimal sequential decision-making algorithms in learning and control
• Top-$k$ eXtreme Contextual Bandits with Arm Hierarchy