arrow
返回

Attentive multi-view reinforcement learning

delete2020-05-04
delete6
PRE
AI
S
Shiliang Sun *
徐鑫 封面图
徐鑫 (Xin Xu)
赵
赵静 (Jing Zhao)
DOI:10.1007/s13042-020-01130-6delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
The reinforcement learning process usually takes millions of steps from scratch, due to the limited observation experience. More precisely, the representation approximated by a single deep network is usually limited for reinforcement learning agents. In this paper, we propose a novel multi-view deep attention network (MvDAN), which introduces multi-view representation learning into the reinforcement learning framework for the first time. Based on the multi-view scheme of function approximation, the proposed model approximates multiple view-specific policy or value functions in parallel by estimating the middle-level representation and integrates these functions based on attention mechanisms to generate a comprehensive strategy. Furthermore, we develop the multi-view generalized policy improvement to jointly optimize all policies instead of a single one. Compared with the single-view function approximation scheme in reinforcement learning methods, experimental results on eight Atari benchmarks show that MvDAN outperforms the state-of-the-art methods and has faster convergence and training stability.
Keyword:
Deep reinforcement learning
Function approximation
Multi-view learning
Representation learning

期刊

International Journal of Machine Learning and Cybernetics 封面图
International Journal of Machine Learning and Cybernetics
IF:
2.7
论文数:
3.2K
被引数:
5.6K

机构

E
east china normal university
学者数:
3.1W
论文数: 2.1W
被引数: 25
N
national university of defense technology - china
学者数:
1.8W
论文数: 1.4W
被引数: 9
引用论文

引用论文

Analysis on smart material suitable for autogenous microelectronic application
err2019-08-30
err0
PREAI
errR Sitharthan; Manimaran Ponnusamy; Madurakavi Karthikeyan; D Shanmuga Sundar
err分享
err收藏
Dynamics of a fringe mangrove forest detected by Landsat images in the Mekong River Delta, Vietnam
err2016-07-04
err0
PREAI
errWilliam Nardin; Silvia Locatelli; Valerie Pasquarella; Maria Cristina Rulli; Curtis E. Woodcock; Sergio Fagherazzi
err分享
err收藏
Multi-view learning overview: Recent progress and new challenges
err2017-11-01
err700
PREAI
errZhao, Jing; Xie, Xijiong; Xu, Xin; Sun, Shiliang
err分享
err收藏
学者 查看更多内容