Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

A comparative investigation on subspace dimension determination.

Xuelei Hu1, Lei Xu

  • 1Department of Computer Science and Engineering, The Chinese University of Hong Kong, Shatin, NT, Hong Kong, China. xlhu@cse.cuhk.edu.hk

Neural Networks : the Official Journal of the International Neural Network Society
|November 24, 2004
PubMed
Summary

This study compares Bayesian Ying-Yang (BYY) harmony learning criteria with traditional methods for determining principal subspace dimensions. BYY criteria demonstrate superior or comparable performance, offering automatic dimension determination.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Topological phase transition induced by spin-orbit coupling in bilayer graphene.

Journal of physics. Condensed matter : an Institute of Physics journal·2013
Same author

[Survey on the occupational musculoskeletal disorder and its risk factors among male steelworkers].

Zhonghua yu fang yi xue za zhi [Chinese journal of preventive medicine]·2013
Same author

Characterisation and identification of dihydroindole-type alkaloids from processed semen strychni by high-performance liquid chromatography coupled with electrospray ionisation ion trap time-of-flight mass spectrometry.

Phytochemical analysis : PCA·2013
Same author

Identification and effect decomposition of risk factors for Brucella contamination of raw whole milk in china.

PloS one·2013
Same author

Different toxicity of the novel Bacillus thuringiensis (Bacillales: Bacillaceae) strain LLP29 against Aedes albopictus and Culex quinquefasciatus (Diptera: Culicidae).

Journal of economic entomology·2013
Same author

Study on the growth and the photosynthetic characteristics of low energy C(+) ion implantation on peanut.

PloS one·2013

Area of Science:

  • Computational Neuroscience
  • Machine Learning
  • Statistical Learning Theory

Background:

  • Constrained Hebbian self-organization identifies principal subspaces, akin to Principal Component Analysis (PCA).
  • Traditional methods often require pre-defined or heuristically determined subspace dimensions (k).
  • Existing statistical model selection criteria include AIC, CAIC, BIC, and CV.

Purpose of the Study:

  • To investigate and compare the performance of Bayesian Ying-Yang (BYY) harmony learning criteria for subspace dimension determination against established criteria.
  • To evaluate the effectiveness of BYY criteria in automatically determining the principal subspace dimension.

Main Methods:

  • Comparative experimental analysis of BYY criteria against AIC, CAIC, BIC, and CV.

Related Experiment Videos

  • Experiments conducted on simulated datasets with varying parameters (sample size, noise variance, dimensions).
  • Validation on real-world datasets from air pollution and sports tracking.

    • Main Results:

    • The Bayesian Inference Criterion (BIC) outperformed AIC, CAIC, and CV.
    • BYY criteria showed performance comparable to or better than BIC.
    • BYY harmony learning automatically determined the subspace dimension during implementation.

    Conclusions:

    • BYY harmony learning is a preferred method for principal subspace dimension determination.
    • BYY offers an advantage by integrating dimension selection within the learning process, unlike traditional methods requiring a two-stage approach.