Alleviating tuning sensitivity in Approximate Dynamic Programming.

Paul Beuchat Angelos Georghiou John Lygeros

Published in: ECC (2016)

Keyphrases

approximate dynamic programming
linear program
dynamic programming
stochastic dynamic programming
reinforcement learning
step size
linear programming
sensitivity analysis
control policy
average cost
factored mdps
policy iteration
decision making
image sequences
optical flow
multi objective