Modeling Explicit Concerning States for Reinforcement Learning in Visual Dialogue.

Zipeng Xu Fandong Meng Xiaojie Wang Duo Zheng Chenxu Lv Jie Zhou

Published in: CoRR (2021)

Keyphrases

reinforcement learning
function approximation
data sets
learning algorithm
high level
natural language
visual information
dialogue system
mixed initiative
low level
visual features
optimal policy
human machine
perceptual aliasing