Illustrated review of convergence conditions of the value iteration algorithm and the rolling horizon procedure for average-cost MDPs

DOI10.1007/s10479-012-1070-0zbMath1251.90381OpenAlexW2026008564MaRDI QIDQ1761758

Eugenio Della Vecchia, Alain Jean-Marie, Silvia C. Di Marco

Publication date: 15 November 2012

Published in: Annals of Operations Research (Search for Journal in Brave)

Full work available at URL: https://hal.inria.fr/inria-00617271/file/RR-7710.pdf

zbMATH Keywords

Markov decision problemsvalue iterationrolling horizonheuristic methods

Mathematics Subject Classification ID

Approximation methods and heuristics in mathematical programming (90C59) Markov and semi-Markov decision processes (90C40)

Related Items

Open Problem—Convergence and Asymptotic Optimality of the Relative Value Iteration in Ergodic Control

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• A simple technique in Markovian control with applications to resource allocation to resource allocation in communication networks
• Contraction mappings underlying undiscounted Markov decision problems
• Iterative solution of the functional equations of undiscounted Markov renewal programming
• Finite state Markovian decision processes
• Limiting Average Criteria For Nonstationary Markov Decision Processes
• Markov Chains and Stochastic Stability
• Rolling Horizon Procedures in Nonhomogeneous Markov Decision Processes
• Geometric convergence of value-iteration in multichain Markov decision problems
• The Asymptotic Behavior of Undiscounted Value Iteration in Markov Decision Problems
• Error bounds for rolling horizon policies in discrete-time Markov control processes
• Étude asymptotique des systèmes markoviens à commande