A Novel Approach for Sampling in Approximate Dynamic Programming Based on F-Discrepancy.

Cristiano Cervellera Danilo Macciò

Published in: IEEE Trans. Cybern. (2017)

Keyphrases

approximate dynamic programming
linear program
stochastic dynamic programming
reinforcement learning
dynamic programming
step size
control policy
factored mdps
policy iteration
linear programming
cost function
monte carlo
average cost
data mining
sample size
global optimization