BriefGPT.xyz
May, 2018
面向任务的多模态层次强化学习策略：视觉对话
Multimodal Hierarchical Reinforcement Learning Policy for Task-Oriented Visual Dialog
HTML
PDF
Jiaping Zhang, Tiancheng Zhao, Zhou Yu
TL;DR
通过多模态层次强化学习框架和状态适应技术，实现了语音和图像的目标驱动对话。在图像猜谜游戏中进行了实验，取得了良好的效果。
Abstract
Creating an
intelligent conversational system
that understands vision and language is one of the ultimate goals in Artificial Intelligence (AI)~\cite{winograd1972understanding}. Extensive research has focused on
vision-
→