ITP OpenIR  > SCI期刊论文
Zhou, HJ
Active Online Learning in the Binary Perceptron Problem
Source PublicationCOMMUNICATIONS IN THEORETICAL PHYSICS
Language英语
KeywordSTATISTICAL-MECHANICS EXAMPLES INFERENCE ALGORITHM NETWORKS STORAGE
AbstractThe binary perceptron is the simplest artificial neural network formed by N input units and one output unit, with the neural states and the synaptic weights all restricted to +/- 1 values. The task in the teacher-student scenario is to infer the hidden weight vector by training on a set of labeled patterns. Previous efforts on the passive learning mode have shown that learning from independent random patterns is quite inefficient. Here we consider the active online learning mode in which the student designs every new Ising training pattern. We demonstrate that it is mathematically possible to achieve perfect (error-free) inference using only N designed training patterns, but this is computationally unfeasible for large systems. We then investigate two Bayesian statistical designing protocols, which require 2.3N and 1.9N training patterns, respectively, to achieve error-free inference. If the training patterns are instead designed through deductive reasoning, perfect inference is achieved using N+log(2)N samples. The performance gap between Bayesian and deductive designing strategies may be shortened in future work by taking into account the possibility of ergodicity breaking in the version space of the binary perceptron.
2019
ISSN0253-6102
Volume71Issue:2Pages:243-252
Subject AreaPhysics
MOST Discipline CataloguePhysics, Multidisciplinary
DOI10.1088/0253-6102/71/2/243
Indexed BySCIE
Citation statistics
Document Type期刊论文
Identifierhttp://ir.itp.ac.cn/handle/311006/23497
CollectionSCI期刊论文
Affiliation1.Chinese Acad Sci, Inst Theoret Phys, CAS Key Lab Theoret Phys, Beijing 100190, Peoples R China
2.Univ Chinese Acad Sci, Sch Phys Sci, Beijing 100049, Peoples R China
Recommended Citation
GB/T 7714
Zhou, HJ. Active Online Learning in the Binary Perceptron Problem[J]. COMMUNICATIONS IN THEORETICAL PHYSICS,2019,71(2):243-252.
APA Zhou, HJ.(2019).Active Online Learning in the Binary Perceptron Problem.COMMUNICATIONS IN THEORETICAL PHYSICS,71(2),243-252.
MLA Zhou, HJ."Active Online Learning in the Binary Perceptron Problem".COMMUNICATIONS IN THEORETICAL PHYSICS 71.2(2019):243-252.
Files in This Item:
There are no files associated with this item.
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Zhou, HJ]'s Articles
Baidu academic
Similar articles in Baidu academic
[Zhou, HJ]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Zhou, HJ]'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.