Monte Carlo TD(lambda)-methods for the optimal control of discrete-time Markovian jump linear systems.

Oswaldo L. V. Costa Julio C. C. Aya

Published in: Autom. (2002)

Keyphrases

monte carlo
optimal control
markov chain
linear systems
monte carlo simulation
monte carlo methods
temporal difference
control problems
reinforcement learning
sparse linear systems
dynamic programming
reinforcement learning methods
matrix inversion
evaluation function
point processes
state space
quasi monte carlo