Combined Reinforcement Learning via Abstract Representations.

Vincent François-Lavet Yoshua Bengio Doina Precup Joelle Pineau

Published in: AAAI (2019)

Keyphrases

reinforcement learning
higher level
function approximation
low level
bayesian networks
neural network
transition model
case study
high level
reinforcement learning algorithms
markov decision processes
optimal control
dynamic programming
temporal difference
probability distribution
multi agent
learning algorithm