An actor-critic algorithm with function approximation for discounted cost constrained Markov decision processes

DOI10.1016/j.sysconle.2010.08.013zbMath1209.90344OpenAlexW2161270100MaRDI QIDQ616967

Shalabh Bhatnagar

Publication date: 12 January 2011

Published in: Systems \& Control Letters (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1016/j.sysconle.2010.08.013

zbMATH Keywords

function approximationactor-critic algorithmconstrained Markov decision processesinfinite horizon discounted cost criterionsimultaneous perturbation stochastic approximation

Mathematics Subject Classification ID

Markov and semi-Markov decision processes (90C40)

Related Items

Constrained Markov decision processes with first passage criteria, Dimension reduction based adaptive dynamic programming for optimal control of discrete-time nonlinear control-affine systems, Risk-Sensitive Reinforcement Learning via Policy Gradient Search, Variance-constrained actor-critic algorithms for discounted and average reward MDPs, Recent advances in reinforcement learning in finance, An online actor-critic algorithm with function approximation for constrained Markov decision processes, Global Convergence of Policy Gradient Methods to (Almost) Locally Optimal Policies, Smoothed functional-based gradient algorithms for off-policy reinforcement learning: a non-asymptotic viewpoint, Risk-Constrained Reinforcement Learning with Percentile Risk Criteria, The Borkar-Meyn theorem for asynchronous stochastic approximations

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Natural actor-critic algorithms
• An actor-critic algorithm for constrained Markov decision processes
• Optimal flow control of a class of queueing networks in equilibrium
• Multivariate stochastic approximation using a simultaneous perturbation gradient approximation
• An analysis of temporal-difference learning with function approximation
• OnActor-Critic Algorithms
• A Kiefer-Wolfowitz algorithm with randomized differences
• The O.D.E. Method for Convergence of Stochastic Approximation and Reinforcement Learning
• Perturbation theory and finite Markov chains