Quasi-Monte Carlo methods for Markov chains with continuous multi-dimensional state space.
Rany El HaddadChristian LécotPierre L'EcuyerN. NassifPublished in: Math. Comput. Simul. (2010)
Keyphrases
- markov chain
- monte carlo methods
- state space
- multi dimensional
- monte carlo
- monte carlo method
- action space
- steady state
- transition probabilities
- reinforcement learning
- finite state
- stationary distribution
- markov process
- random walk
- dynamic programming
- markov decision processes
- markov model
- particle filter
- transition matrix
- markov processes
- optimal policy
- search space
- initial state
- markov decision process
- reward function
- markov random field
- probabilistic automata