A General Family of Robust Stochastic Operators for Reinforcement Learning

2018/05/21 by Yingdong Lu, Lu, Yingdong, Mark S. Squillante +3 · 1 citation
Computer Science · #Adaptive Dynamic Programming Control #Evolutionary Algorithms and Applications #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Reinforcement Learning in Robotics

paper · pdf · doi:10.48550/arxiv.1805.08122

openalex publication_date 2018/05/21 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We consider a new family of operators for reinforcement learning with the goal of alleviating the negative effects and becoming more robust to approximation or estimation errors. Various theoretical results are established, which include showing on a sample path basis that our family of operators preserve optimality and increase the action gap. Our empirical results illustrate the strong benefits of our family of operators, significantly outperforming the classical Bellman operator and recently proposed operators.

Cited by

• Bellman operator convergence enhancements in reinforcement learning algorithms

Related

• On Practical Reinforcement Learning: Provable Robustness, Scalability, and Statistical Efficiency
• Robust Reinforcement Learning with Distributional Risk-averse formulation
• Proximal Bellman mappings for reinforcement learning and their application to robust adaptive filtering
• Improving reinforcement learning algorithms: towards optimal learning rate policies
• A Tutorial Introduction to Reinforcement Learning