A unified approach to Markov decision problems and performance sensitivity analysis with discounted and average criteria: multichain cases

DOI10.1016/J.AUTOMATICA.2004.05.003zbMath1059.90141OpenAlexW2051442802MaRDI QIDQ705478

Xianping Guo, Cao, Xiren

Publication date: 31 January 2005

Published in: Automatica (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1016/j.automatica.2004.05.003

zbMATH Keywords

PotentialsReinforcement learningPerformance sensitivityPerturbation analysisPolicy iteration

Mathematics Subject Classification ID

Sensitivity (robustness) (93B35) Optimal stochastic control (93E20) Markov and semi-Markov decision processes (90C40)

Related Items (7)

The risk probability criterion for discounted continuous-time Markov decision processes ⋮ Optimization of a special case of continuous-time Markov decision processes with compact action set ⋮ Completion-of-squares: revisited and extended ⋮ Temporal difference-based policy iteration for optimal control of stochastic systems ⋮ Continuous-time Markov decision processes with \(n\)th-bias optimality criteria ⋮ Basic ideas for event-based optimization of Markov systems ⋮ Bias optimality for multichain continuous-time Markov decision processes

Cites Work

• A Brouwer fixed-point mapping approach to communicating Markov decision processes
• Foolproof convergence in multichain policy iteration
• Single sample path-based optimization of Markov chains
• Realization probabilities. The dynamics of queuing systems
• The relations among potentials, perturbation analysis, and Markov decision processes
• Minimax control for discrete-time time-varying stochastic systems
• From perturbation analysis to Markov decision processes and reinforcement learning
• A note on policy algorithms for discounted Markov decision problems
• A unified approach to Markov decision problems and performance sensitivity analysis
• Limiting Average Criteria For Nonstationary Markov Decision Processes
• A Fixed Point Approach to Undiscounted Markov Renewal Programs
• The Maclaurin series for performance functions of Markov chains
• Perturbation realization, potentials, and sensitivity analysis of Markov processes
• CONVERGENCE OF SIMULATION-BASED POLICY ITERATION
• Simulation-based optimization of Markov reward processes
• Discrete Dynamic Programming
• On Finding Optimal Policies in Discrete Dynamic Programming with No Discounting
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item

This page was built for publication: A unified approach to Markov decision problems and performance sensitivity analysis with discounted and average criteria: multichain cases