Approximate dynamic programming, local or global optimal solution?

Ali Heydari Sivasubramanya Nadar Balakrishnan

Published in: ACC (2014)

Keyphrases

approximate dynamic programming
linear program
optimal solution
linear programming
stochastic dynamic programming
dynamic programming
reinforcement learning
objective function
factored mdps
np hard
search space
policy iteration
data mining
lower bound
multistage
step size