Average-Reward Learning and Planning with Options

2021/10/26 by Yi Wan, Abhishek Naik, Wan, Yi +3 · 1 citation
Biochemistry, Genetics and Molecular Biology · Computer Science · Decision Sciences · #Advanced Bandit Algorithms Research #FOS: Computer and information sciences #Machine Learning (cs.LG) #Receptor Mechanisms and Signaling #Reinforcement Learning in Robotics

paper · pdf · doi:10.48550/arxiv.2110.13855

openalex publication_date 2021/10/26 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We extend the options framework for temporal abstraction in reinforcement learning from discounted Markov decision processes (MDPs) to average-reward MDPs. Our contributions include general convergent off-policy inter-option learning algorithms, intra-option algorithms for learning values and models, as well as sample-based planning variants of our learning algorithms. Our algorithms and convergence proofs extend those recently developed by Wan, Naik, and Sutton. We also extend the notion of option-interrupting behavior from the discounted to the average-reward formulation. We show the efficacy of the proposed algorithms with experiments on a continuing version of the Four-Room domain.

Citations

• Stochastic Approximation: A Dynamical Systems Viewpoint
• Between MDPs and semi-MDPs: A framework for temporal abstraction in reinforcement learning

Cited by

• Average-reward reinforcement learning in semi-Markov decision processes via relative value iteration

Related

• Learning and Planning in Average-Reward Markov Decision Processes
• The Option-Critic Architecture
• Hierarchical Average Reward Policy Gradient Algorithms
• Reinforcement Learning for Infinite-Horizon Average-Reward Linear MDPs via Approximation by Discounted-Reward MDPs
• Average Reward Adjusted Discounted Reinforcement Learning: Near-Blackwell-Optimal Policies for Real-World Applications