On the Convergence of Policy Iteration in Stationary Dynamic Programming

DOI10.1287/moor.4.1.60zbMath0411.90072MaRDI QIDQ4198357

Shelby Brumelle, Martin L. Puterman

Publication date: 1979

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

Full work available at URL: https://doi.org/10.1287/moor.4.1.60

zbMATH Keywords

rate of convergence; error bounds; compact action space; programming in abstract spaces; policy iteration method; partially ordered normed linear spaces; finite state Markovian decision problem; Newton-Kantorovich iteration procedure; stationary dynamic programming

Mathematics Subject Classification ID

90C47: Minimax problems in mathematical programming

49M15: Newton-type methods

46B99: Normed linear spaces and Banach spaces; Banach lattices

90C39: Dynamic programming

90C48: Programming in abstract spaces

90C40: Markov and semi-Markov decision processes

41A25: Rate of convergence, degree of approximation

Related Items

Survey of linear programming for standard and nonstandard Markovian control problems. Part I: Theory, Multigrid methods for two‐player zero‐sum stochastic games, NUMERICAL METHODS FOR DIFFERENTIAL GAMES BASED ON PARTIAL DIFFERENTIAL EQUATIONS, (Approximate) iterated successive approximations algorithm for sequential decision processes, A numerical method for pricing European options with proportional transaction costs, The variational calculus and approximation in policy space for Markovian decision processes, Discrete dynamic programming and viscosity solutions of the Bellman equation, On the convergence of policy iteration for controlled diffusions, Policy iteration and Newton-Raphson methods for Markov decision processes under average cost criterion, Unnamed Item, A Fixed Point Approach to Undiscounted Markov Renewal Programs