Constrained Policy Improvement for Efficient Reinforcement Learning.

Elad Sarafian Aviv Tamar Sarit Kraus

Published in: IJCAI (2020)

Keyphrases

reinforcement learning
optimal policy
computationally expensive
neural network
lightweight
function approximation
policy search
action selection
policy iteration
cost effective
markov decision processes
markov decision process
function approximators
control policy
partially observable environments