Agent-Aware Dropout DQN for Safe and Efficient On-line Dialogue Policy Learning.

Lu Chen Xiang Zhou Cheng Chang Runzhe Yang Kai Yu

Published in: EMNLP (2017)

Keyphrases

learning algorithm
learning systems
practical reasoning
knowledge level
intelligent agents
multi agent
learning process
multi agent systems
learning mechanism
prior knowledge
online learning
knowledge acquisition
learning tasks
multiple agents
partially observable
learning agent
decision making