Attention-based Deep Reinforcement Learning for Multi-view Environments

被引:0
|
作者
Barati, Elaheh [1 ]
Chen, Xuewen [2 ]
Zhong, Zichun [1 ]
机构
[1] Wayne State Univ, Detroit, MI 48202 USA
[2] AIWAYS AUTO, Shanghai, Peoples R China
关键词
Reinforcement learning; Deep learning; Attention networks;
D O I
暂无
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
In reinforcement learning algorithms, it is a common practice to account for only a single view of the environment to make the desired decisions; however, utilizing multiple views of the environment can help to promote the learning of complicated policies. Since the views may frequently suffer from partial observability, their provided observation can have different levels of importance. In this paper, we present a novel attention-based deep reinforcement learning method in a multi-view environment in which each view can provide various representative information about the environment. Specifically, our method learns a policy to dynamically attend to views of the environment based on their importance in the decision-making process. We evaluate the performance of our method on TORCS racing car simulator and three other complex 3D environments with obstacles.
引用
下载
收藏
页码:1805 / 1807
页数:3
相关论文
共 50 条
  • [31] Deep Co-Attention Network for Multi-View Subspace Learning
    Zheng, Lecheng
    Cheng, Yu
    Yang, Hongxia
    Cao, Nan
    He, Jingrui
    PROCEEDINGS OF THE WORLD WIDE WEB CONFERENCE 2021 (WWW 2021), 2021, : 1528 - 1539
  • [32] Attention-Based Deep Reinforcement Learning for Virtual Cinematography of 360° Videos
    Wang, Jianyi
    Xu, Mai
    Jiang, Lai
    Song, Yuhang
    IEEE TRANSACTIONS ON MULTIMEDIA, 2021, 23 : 3227 - 3238
  • [33] Attention-based network for effective action recognition from multi-view video
    Hoang-Thuyen Nguyen
    Thi-Oanh Nguyen
    KNOWLEDGE-BASED AND INTELLIGENT INFORMATION & ENGINEERING SYSTEMS (KSE 2021), 2021, 192 : 971 - 980
  • [34] ATTENTION-BASED CURIOSITY-DRIVEN EXPLORATION IN DEEP REINFORCEMENT LEARNING
    Reizinger, Patrik
    Szemenyei, Marton
    2020 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, 2020, : 3542 - 3546
  • [35] An attention-based and deep sparse priori cascade multi-view stereo network for 3D reconstruction
    Wang, Yadong
    Ran, Teng
    Liang, Yuan
    Zheng, Guoquan
    COMPUTERS & GRAPHICS-UK, 2023, 116 : 383 - 392
  • [36] Multi-View Object Detection Based on Deep Learning
    Tang, Cong
    Ling, Yongshun
    Yang, Xing
    Jin, Wei
    Zheng, Chao
    APPLIED SCIENCES-BASEL, 2018, 8 (09):
  • [37] Multi-Task Reinforcement Learning With Attention-Based Mixture of Experts
    Cheng, Guangran
    Dong, Lu
    Cai, Wenzhe
    Sun, Changyin
    IEEE ROBOTICS AND AUTOMATION LETTERS, 2023, 8 (06) : 3811 - 3818
  • [38] Generalized attention-based deep multi-instance learning
    Zhao, Lu
    Yuan, Liming
    Hao, Kun
    Wen, Xianbin
    MULTIMEDIA SYSTEMS, 2023, 29 (01) : 275 - 287
  • [39] Generalized attention-based deep multi-instance learning
    Lu Zhao
    Liming Yuan
    Kun Hao
    Xianbin Wen
    Multimedia Systems, 2023, 29 : 275 - 287
  • [40] Predicting drug-drug interactions based on multi-view and multichannel attention deep learning
    Huang, Liyu
    Chen, Qingfeng
    Lan, Wei
    HEALTH INFORMATION SCIENCE AND SYSTEMS, 2023, 11 (01)