An Incremental Fast Policy Search Using a Single Sample Path

DOI10.1007/978-3-319-69900-4_1MaRDI QIDQ5045345zbMATH OpenOpenAlexFDO

Authors Ajin George Joseph, Shalabh Bhatnagar

Publication date 4 November 2022

Published in Lecture Notes in Computer Science (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1007/978-3-319-69900-4_1

Mathematics Subject Classification ID

Applications of Markov chains and discrete-time Markov processes on general state spaces (social mobility, learning theory, industrial processes, etc.) (60J20) Learning and adaptive systems in artificial intelligence (68T05) Markov and semi-Markov decision processes (90C40)

Recommendations

• An incremental off-policy search in a model-free Markov decision process using a single sample path
• Learning near-optimal policies with Bellman-residual minimization based fitted policy iteration and a single sample path
• Learning Near-Optimal Policies with Bellman-Residual Minimization Based Fitted Policy Iteration and a Single Sample Path
• Sample-based policy iteration for constrained DEC-POMDPs
• Efficient sample reuse in policy gradients with parameter-based exploration
• Rollout sampling approximate policy iteration
• On-line policy gradient estimation with multi-step sampling

Cites work

• An iterative method of solving a game
• Basis function adaptation in temporal difference reinforcement learning
• Cross-entropy and rare events for maximal cut and partition problems
• scientific article; zbMATH DE number 5685899 (Why is no real title available?)
• Revisiting the cross entropy method with applications in stochastic global optimization and reinforcement learning
• The Complexity of Markov Decision Processes
• The cross-entropy method for combinatorial and continuous optimization
• The cross-entropy method for continuous multi-extremal optimization

Cited in

(6)

• Learning near-optimal policies with Bellman-residual minimization based fitted policy iteration and a single sample path
• An incremental off-policy search in a model-free Markov decision process using a single sample path
• A generalized path integral control approach to reinforcement learning
• scientific article; zbMATH DE number 6860770 (Why is no real title available?)
• Adaptive smoothing for path integral control
• Learning Near-Optimal Policies with Bellman-Residual Minimization Based Fitted Policy Iteration and a Single Sample Path

This page was built for publication: An Incremental Fast Policy Search Using a Single Sample Path

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5045345)