"I'm Sorry Dave, I'm Afraid I Can't Do That" Deep Q-Learning from Forbidden Actions.

Mathieu Seurin Philippe Preux Olivier Pietquin

Published in: IJCNN (2020)

Keyphrases

action selection
reinforcement learning
function approximation
state action space
cooperative
learning algorithm

state space
multi agent
reasoning about actions
state action
learning agent
goal directed

reward function
plan recognition
stochastic approximation
temporal difference
optimal policy
data sets

initial state
decision theoretic
situation calculus
decision making
bucket brigade
neural network