Reinforcement learning based algorithms for average cost Markov decision processes

DOI10.1007/s10626-006-0003-yzbMath1146.90521OpenAlexW2061769118MaRDI QIDQ2643632

Mohammed Shahid Abdulla, Shalabh Bhatnagar

Publication date: 27 August 2007

Published in: Discrete Event Dynamic Systems (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1007/s10626-006-0003-y

zbMATH Keywords

Markov decision processesreinforcement learningpolicy iterationactor-critic algorithmssimultaneous perturbation stochastic approximationnormalized Hadamard matricesTD-learningtwo timescale stochastic approximation

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Markov and semi-Markov decision processes (90C40)

Related Items

A constrained optimization perspective on actor-critic algorithms and application to network routing, Multiscale Q-learning with linear function approximation, Natural actor-critic algorithms

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• A one-measurement form of simultaneous perturbation stochastic approximation
• Dynamic programming and stochastic control
• Actor-critic algorithms for hierarchical Markov decision processes
• Average cost temporal-difference learning
• Multivariate stochastic approximation using a simultaneous perturbation gradient approximation
• An analysis of temporal-difference learning with function approximation
• Asynchronous Stochastic Approximations
• OnActor-Critic Algorithms
• Actor-Critic--Type Learning Algorithms for Markov Decision Processes
• The O.D.E. Method for Convergence of Stochastic Approximation and Reinforcement Learning
• A Simultaneous Perturbation Stochastic Approximation-Based Actor–Critic Algorithm for Markov Decision Processes
• The actor-critic algorithm as multi-time-scale stochastic approximation.