摘要: Deep Recurrent Q-Learning for Partially Observable MDPs 摘要:DQN 的两个缺陷,分别是:limited memory 和 rely on being able to perceive the complete game screen at e 阅读全文
posted @ 2016-10-03 21:25 AHU-WangXiao 阅读(4349) 评论(0) 推荐(0) 编辑
摘要: Main Menu Fortune.com Main Menu Fortune.com Fortune.com Fortune.com E-mail Tweet Facebook Linkedin Share icons By Roger Parloff Illustration by Justin 阅读全文
posted @ 2016-10-03 17:06 AHU-WangXiao 阅读(900) 评论(0) 推荐(0) 编辑
摘要: Deep Attention Recurrent Q-Network 5vision groups 摘要:本文将 DQN 引入了 Attention 机制,使得学习更具有方向性和指导性。(前段时间做一个工作打算就这么干,谁想到,这么快就被这几个孩子给实现了,自愧不如啊( ⊙ o ⊙ )) 引言:我们 阅读全文
posted @ 2016-10-03 15:34 AHU-WangXiao 阅读(3650) 评论(0) 推荐(0) 编辑