中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 自动化研究所
  3. 中国科学院自动化研究所
  4. 脑图谱与类脑智能实验室
  5. 类脑认知计算
Solving the spike feature information vanishing problem in spiking deep Q network with potential based normalization

文献类型：期刊论文

作者Sun, Yinqian1,5; Zeng, Yi1,2,3,4,5; Li, Yang1,4
刊名FRONTIERS IN NEUROSCIENCE
出版日期2022-08-25
卷号16页码:11
关键词brain-inspired decision model SDQN reinforcement learning potential normalization spiking activity
DOI10.3389/fnins.2022.953368
通讯作者Zeng, Yi(yi.zeng@ia.ac.cn)
英文摘要Brain-inspired spiking neural networks (SNNs) are successfully applied to many pattern recognition domains. The SNNs-based deep structure has achieved considerable results in perceptual tasks, such as image classification and target detection. However, applying deep SNNs in reinforcement learning (RL) tasks is still a problem to be explored. Although there have been previous studies on the combination of SNNs and RL, most focus on robotic control problems with shallow networks or using the ANN-SNN conversion method to implement spiking deep Q networks (SDQN). In this study, we mathematically analyzed the problem of the disappearance of spiking signal features in SDQN and proposed a potential-based layer normalization (pbLN) method to train spiking deep Q networks directly. Experiment shows that compared with state-of-art ANN-SNN conversion method and other SDQN works, the proposed pbLN spiking deep Q networks (PL-SDQN) achieved better performance on Atari game tasks.
资助项目National Key Research and Development Program ; Strategic Priority Research Program of the Chinese Academy of Sciences ; National Natural Science Foundation of China ; [2020AAA0104305] ; [XDB32070100] ; [62106261]
WOS研究方向Neurosciences & Neurology
语种英语
WOS记录号WOS:000852629100001
出版者FRONTIERS MEDIA SA
资助机构National Key Research and Development Program ; Strategic Priority Research Program of the Chinese Academy of Sciences ; National Natural Science Foundation of China
源URL[http://ir.ia.ac.cn/handle/173211/50083]  
专题类脑智能研究中心_类脑认知计算
通讯作者Zeng, Yi
作者单位1.Chinese Acad Sci, Inst Automat, Res Ctr Brain Inspired Intelligence, Beijing, Peoples R China
2.Chinese Acad Sci, Ctr Excellence Brain Sci & Intelligence Technol, Shanghai, Peoples R China
3.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing, Peoples R China
4.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing, Peoples R China
5.Univ Chinese Acad Sci, Sch Future Technol, Beijing, Peoples R China
推荐引用方式
GB/T 7714		 Sun, Yinqian,Zeng, Yi,Li, Yang. Solving the spike feature information vanishing problem in spiking deep Q network with potential based normalization[J]. FRONTIERS IN NEUROSCIENCE,2022,16:11.
APA Sun, Yinqian,Zeng, Yi,&Li, Yang.(2022).Solving the spike feature information vanishing problem in spiking deep Q network with potential based normalization.FRONTIERS IN NEUROSCIENCE,16,11.
MLA Sun, Yinqian,et al."Solving the spike feature information vanishing problem in spiking deep Q network with potential based normalization".FRONTIERS IN NEUROSCIENCE 16(2022):11.

入库方式： OAI收割

来源：自动化研究所

浏览0
下载0
收藏0
其他版本

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。