Some basic concepts of numerical treatment of Markov decision models
From MaRDI portal
The authors review some basic results on Markov decision processes. Focus is on numerical aspects. Applied to different optimality criteria they outline the classical methods for calculating the value function and/or an optimal policy and consider methods of discretization, extrapolation, and, finally, methods for detecting suboptimal actions.
Recommendations
- scientific article; zbMATH DE number 4039674
- Numerical approximations for discounted continuous time Markov decision processes
- Numerical Methods in Markov Chain Modeling
- Numerical analysis of continuous time Markov decision processes over finite horizons
- Numerical analysis of large Markov reward models
- scientific article; zbMATH DE number 794375
- scientific article; zbMATH DE number 729460
- Markov decision processes with their applications
- Markov and Markov reward model transient analysis: An overview of numerical approaches
- Markov decision processes in practice
Cites work
- A decision exclusion algorithm for a class of Markovian Decision Processes
- A method of bisection for discounted Markov decision problems
- A modified dynamic programming method for Markovian decision problems
- A modified form of the iterative method of dynamic programming
- A successive approximation algorithm for an undiscounted Markov decision process
- Approximations of Dynamic Programs, I
- Convergence of discretization procedures in dynamic programming
- Discounted Dynamic Programming
- Discrete Dynamic Programming
- Dynamic programming, Markov chains, and the method of successive approximations
- Estimates for finite-stage dynamic programs
- Finite-state approximations to denumerable-state dynamic programs
- scientific article; zbMATH DE number 3148886 (Why is no real title available?)
- scientific article; zbMATH DE number 3864994 (Why is no real title available?)
- scientific article; zbMATH DE number 3460142 (Why is no real title available?)
- scientific article; zbMATH DE number 3628710 (Why is no real title available?)
- Linear programming and sequential decisions
- Multichain Markov Renewal Programs
- Multiple Policy Improvements in Undiscounted Markov Renewal Programming
- Negative Dynamic Programming
- Note—A Test for Nonoptimal Actions in Undiscounted Finite Markov Decision Chains
- On Finding the Maximal Gain for Markov Decision Processes
- On sequential decisions and Markov chains
- On the Fixed Points of the Optimal Reward Operator in Stochastic Dynamic Programming with Discount Factor Greater than One
- On the Opimality of $( {s,S} )$ Inventory Policies: New Conditions and a New Proof
- Solution of a Markovian decision problem by successive overrelaxation
- Some Bounds for Discounted Sequential Decision Processes
- Technical Note—Bounds on the Gain of a Markov Decision Process
Cited in
(5)- Methods of optimal dynamic decisions using the discount criterion
- scientific article; zbMATH DE number 3906263 (Why is no real title available?)
- scientific article; zbMATH DE number 4039674 (Why is no real title available?)
- Numerical analysis of large Markov reward models
- Suboptimal policy determination for large-scale Markov decision processes. II: Implementation and numerical evaluation
This page was built for publication: Some basic concepts of numerical treatment of Markov decision models
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3743147)