Minimal Residual Approaches for Policy Evaluation in Large Sparse Markov Chains.

Hengshuai Yao Zhi-Qiang Liu

Published in: ISAIM (2008)

Keyphrases

markov chain
steady state
policy evaluation
monte carlo
finite state
transition probabilities
markov processes
least squares
state space
markov model
temporal difference
sample path
dynamic programming
optimal control
importance sampling
markov decision processes
variance reduction
random walk