Q-learning and policy iteration algorithms for stochastic shortest path problems

DOI10.1007/S10479-012-1128-ZzbMath1306.90171OpenAlexW2027855416WikidataQ115147448 ScholiaQ115147448MaRDI QIDQ378731

Huizhen Yu, Dimitri P. Bertsekas

Publication date: 12 November 2013

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

Full work available at URL: https://doi.org/10.1007/s10479-012-1128-z

zbMATH Keywords

Markov decision processesstochastic approximationpolicy iterationstochastic shortest pathsvalue iterationapproximate dynamic programmingQ-learning

Mathematics Subject Classification ID

Dynamic programming (90C39) Markov and semi-Markov decision processes (90C40)

Related Items (5)

A Mixed Value and Policy Iteration Method for Stochastic Control with Universally Measurable Policies ⋮ Error bounds for constant step-size \(Q\)-learning ⋮ Proximal algorithms and temporal difference methods for solving fixed point problems ⋮ Robust shortest path planning and semicontractive dynamic programming ⋮ Fundamental design principles for reinforcement learning algorithms

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• (Approximate) iterated successive approximations algorithm for sequential decision processes
• A generalized Kalman filter for fixed point approximation and efficient temporal-difference learning
• Projected equation methods for approximate solution of large linear systems
• Asynchronous stochastic approximation and Q-learning
• Finite state Markovian decision processes
• Q-Learning and Enhanced Policy Iteration in Discounted Dynamic Programming
• Distributed asynchronous computation of fixed points
• Distributed dynamic programming
• An Analysis of Stochastic Shortest Path Problems
• On Stationary Strategies in Borel Dynamic Programming
• Asynchronous Iterative Methods for Multiprocessors
• On the Convergence of Stochastic Iterative Dynamic Programming Algorithms
• Optimal stopping of Markov processes: Hilbert space theory, approximation algorithms, and an application to pricing high-dimensional financial derivatives
• On Boundedness of Q-Learning Iterates for Stochastic Shortest Path Problems
• Neuro-Dynamic Programming: An Overview and Recent Results
• Discrete Dynamic Programming with Sensitive Discount Optimality Criteria

This page was built for publication: Q-learning and policy iteration algorithms for stochastic shortest path problems