Prioritization methods for accelerating MDP solvers

MaRDI QIDQ3093280zbMATH OpenFDO

Authors D. L. Wingate, Kevin D. Seppi

Publication date 12 October 2011

Full work available at URL http://www.jmlr.org/papers/v6/wingate05a.html

zbMATH Keywords

dynamic programmingvalue iterationMarkov decision processespolicy iterationprioritized sweeping

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05)

Recommendations

• Acceleration Operators in the Value Iteration Algorithms for Markov Decision Processes
• Topological value iteration algorithms
• Policy iteration accelerated with Krylov methods
• Accelerated modified policy iteration algorithms for Markov decision processes
• On iterative optimization ol structured Markov decision processes with discounted rewards

Cited in

(5)

• Robust topological policy iteration for infinite horizon bounded Markov decision processes
• Ranking policies in discrete Markov decision processes
• Policy iteration accelerated with Krylov methods
• Topological value iteration algorithms
• Accelerating the convergence of value iteration by using partial transition functions

Describes a project that uses

Uses Software

• Aztec
• GraphBase

This page was built for publication: Prioritization methods for accelerating MDP solvers

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3093280)