On the existence of fixed points for approximate value iteration and temporal-difference learning

DOI10.1023/A:1004641123405zbMath1028.90077MaRDI QIDQ1586803

D. P. de Farias, Benjamin van Roy

Publication date: 19 February 2001

Published in: Journal of Optimization Theory and Applications (Search for Journal in Brave)

zbMATH Keywords

dynamic programmingreinforcement learningvalue iterationtemporal-difference learningneurodynamic programming

Mathematics Subject Classification ID

Dynamic programming (90C39)

Related Items (8)

Least squares policy iteration with instrumental variables vs. direct policy search: comparison against optimal benchmarks using energy storage ⋮ A perturbation approach to a class of discounted approximate value iteration algorithms with Borel spaces ⋮ A generalized Kalman filter for fixed point approximation and efficient temporal-difference learning ⋮ Unnamed Item ⋮ Shape constraints in economics and operations research ⋮ Application of interval iterations to the entrainment problem in respiratory physiology ⋮ A perturbation approach to approximate value iteration for average cost Markov decision processes with Borel spaces and bounded costs ⋮ Analyzing Approximate Value Iteration Algorithms

Cites Work

• Unnamed Item
• Unnamed Item
• The convergence of \(TD(\lambda)\) for general \(\lambda\)
• Functional Approximations and Dynamic Programming
• An analysis of temporal-difference learning with function approximation

This page was built for publication: On the existence of fixed points for approximate value iteration and temporal-difference learning