Sequential Constant Size Compressors for Reinforcement Learning.

Linus Gisslén Matthew D. Luciw Vincent Graziano Jürgen Schmidhuber

Published in: AGI (2011)

Keyphrases

reinforcement learning
neural network
learning algorithm
function approximation
space complexity
fixed size
real world
machine learning
social networks
state space
markov chain
markov decision processes