Algorithm for Constrained Markov Decision Process with Linear Convergence.
Egor GladinMaksim Lavrik-KarmazinKarina ZainullinaVarvara RudenkoAlexander V. GasnikovMartin TakácPublished in: CoRR (2022)
Keyphrases
- iterative algorithms
- learning algorithm
- markov decision process
- k means
- search space
- dynamic programming
- markov decision processes
- optimal solution
- state space
- convergence rate
- temporal difference learning
- infinite horizon
- markov model
- multistage
- random walk
- em algorithm
- expectation maximization
- probabilistic model
- computational complexity