An incremental off-policy search in a model-free Markov decision process using a single sample path

DOI10.1007/s10994-018-5697-1zbMath1465.90116arXiv1801.10287OpenAlexW2963057120MaRDI QIDQ1621868

Ajin George Joseph, Shalabh Bhatnagar

Publication date: 12 November 2018

Published in: Machine Learning (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1801.10287

zbMATH Keywords

global optimizationMarkov decision processcontrol problemcross entropy methodlinear function approximationstochastic approximation methodODE methodoff-policy prediction

Mathematics Subject Classification ID

Markov and semi-Markov decision processes (90C40)

Uses Software

• PILCO

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Model-based search for combinatorial optimization: A critical survey
• The cross-entropy method for continuous multi-extremal optimization
• Natural actor-critic algorithms
• On diagonal dominance arguments for bounding \(\| A^{-1}\|_\infty\)
• A note on entrywise perturbation theory for Markov chains
• Simulation-based algorithms for Markov decision processes
• The cross-entropy method for combinatorial and continuous optimization
• Application of the cross-entropy method to the buffer allocation problem in a simulation-based environment
• Basis function adaptation in temporal difference reinforcement learning
• Policy Iteration Based on Stochastic Factorization
• Importance Sampling for Stochastic Simulations
• A Model Reference Adaptive Search Method for Global Optimization
• Learning control of finite Markov chains with an explicit trade-off between estimation and control
• Adaptive aggregation methods for infinite horizon dynamic programming
• Optimal adaptive controllers for unknown Markov chains
• Learning control of finite Markov chains with unknown transition probabilities
• Multivariate stochastic approximation using a simultaneous perturbation gradient approximation
• Acceleration of Stochastic Approximation by Averaging
• An analysis of temporal-difference learning with function approximation
• OnActor-Critic Algorithms
• Cross-entropy and rare events for maximal cut and partition problems
• 10.1162/1532443041827907
• Least Squares Temporal Difference Methods: An Analysis under General Conditions
• Learning Near-Optimal Policies with Bellman-Residual Minimization Based Fitted Policy Iteration and a Single Sample Path
• A Stochastic Approximation Framework for a Class of Randomized Optimization Algorithms
• Parameter Estimation for ODEs Using a Cross-Entropy Approach
• Handbook of Markov decision processes. Methods and applications

This page was built for publication: An incremental off-policy search in a model-free Markov decision process using a single sample path