Efficient model-free Q-faetor approximation in value space via log-sum-exp neural networks.

Giuseppe Carlo Calafiore Corrado Possieri

Published in: ECC (2020)

Keyphrases

model free
neural network
reinforcement learning
pattern recognition
temporal difference
space requirements
artificial neural networks
search space
control system
function approximation
policy evaluation
data mining
feature extraction
reinforcement learning algorithms