From perturbation analysis to Markov decision processes and reinforcement learning

DOI10.1023/A:1022188803039zbMath1031.93166MaRDI QIDQ1870309

Xi-Ren Cao

Publication date: 11 May 2003

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

zbMATH Keywords

Markov decision processesperturbation analysisreinforcement learningon-line algorithmsPoisson equationsperformance potentialsQ-learninggradient-based policy iterationTD(\(\lambda\))

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Perturbations in control/observation systems (93C73) Stochastic learning and adaptive control (93E35) Markov and semi-Markov decision processes (90C40)

Related Items (6)

Stochastic control via direct comparison ⋮ Performance optimization of queueing systems with perturbation realization ⋮ Policy iteration based feedback control ⋮ Continuous-time Markov decision processes with \(n\)th-bias optimality criteria ⋮ A unified approach to Markov decision problems and performance sensitivity analysis with discounted and average criteria: multichain cases ⋮ Error bounds of optimization algorithms for semi-Markov decision processes

This page was built for publication: From perturbation analysis to Markov decision processes and reinforcement learning