ITP OpenIR  > SCI期刊论文
Huang, Haiping; Zhou, Haijun; Huang, HP , Chinese Acad Sci, Key Lab Frontiers Theoret Phys, Inst Theoret Phys, Beijing 100190, Peoples R China
Learning by random walks in the weight space of the Ising perceptron
Source PublicationJOURNAL OF STATISTICAL MECHANICS-THEORY AND EXPERIMENT
KeywordConstraint Satisfaction Problems Neural Network Models Binary Perceptron Storage Capacity Synapses Algorithm Couplings Plasticity Dynamics
AbstractSeveral variants of a stochastic local search process for constructing the synaptic weights of an Ising perceptron are studied. In this process, binary patterns are sequentially presented to the Ising perceptron and are then learned as the synaptic weight configuration is modified through a chain of single- or double-weight flips within the compatible weight configuration space of the earlier learned patterns. This process is able to reach a storage capacity of alpha approximate to 0.63 for pattern length N = 101 and alpha approximate to 0.41 for N = 1001. If in addition a relearning process is exploited, the learning performance is further improved to a storage capacity of alpha approximate to 0.80 for N = 101 and alpha approximate to 0.42 for N = 1001. We found that, for a given learning task, the solutions constructed by the random walk learning process are separated by a typical Hamming distance, which decreases with the constraint density a of the learning task; at a fixed value of a, the width of the Hamming distance distribution decreases with N.
2010
ISSN1742-5468
Issue75Pages:-
Subject AreaPhysics
Indexed BySCI
Funding OrganizationNational Science Foundation of China[10774150, 10834014]; China 973-Program[2007CB935903] ; National Science Foundation of China[10774150, 10834014]; China 973-Program[2007CB935903] ; National Science Foundation of China[10774150, 10834014]; China 973-Program[2007CB935903] ; National Science Foundation of China[10774150, 10834014]; China 973-Program[2007CB935903]
Citation statistics
Cited Times:5[WOS]   [WOS Record]     [Related Records in WOS]
Document Type期刊论文
Identifierhttp://ir.itp.ac.cn/handle/311006/5072
CollectionSCI期刊论文
Corresponding AuthorHuang, HP , Chinese Acad Sci, Key Lab Frontiers Theoret Phys, Inst Theoret Phys, Beijing 100190, Peoples R China
Recommended Citation
GB/T 7714
Huang, Haiping,Zhou, Haijun,Huang, HP , Chinese Acad Sci, Key Lab Frontiers Theoret Phys, Inst Theoret Phys, Beijing 100190, Peoples R China. Learning by random walks in the weight space of the Ising perceptron[J]. JOURNAL OF STATISTICAL MECHANICS-THEORY AND EXPERIMENT,2010(75):-.
APA Huang, Haiping,Zhou, Haijun,&Huang, HP , Chinese Acad Sci, Key Lab Frontiers Theoret Phys, Inst Theoret Phys, Beijing 100190, Peoples R China.(2010).Learning by random walks in the weight space of the Ising perceptron.JOURNAL OF STATISTICAL MECHANICS-THEORY AND EXPERIMENT(75),-.
MLA Huang, Haiping,et al."Learning by random walks in the weight space of the Ising perceptron".JOURNAL OF STATISTICAL MECHANICS-THEORY AND EXPERIMENT .75(2010):-.
Files in This Item:
File Name/Size DocType Version Access License
Learning by random w(305KB) 开放获取LicenseApplication Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Huang, Haiping]'s Articles
[Zhou, Haijun]'s Articles
[Huang, HP , Chinese Acad Sci, Key Lab Frontiers Theoret Phys, Inst Theoret Phys, Beijing 100190, Peoples R China]'s Articles
Baidu academic
Similar articles in Baidu academic
[Huang, Haiping]'s Articles
[Zhou, Haijun]'s Articles
[Huang, HP , Chinese Acad Sci, Key Lab Frontiers Theoret Phys, Inst Theoret Phys, Beijing 100190, Peoples R China]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Huang, Haiping]'s Articles
[Zhou, Haijun]'s Articles
[Huang, HP , Chinese Acad Sci, Key Lab Frontiers Theoret Phys, Inst Theoret Phys, Beijing 100190, Peoples R China]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.