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

Neural network learning algorithms for tracking minor subspace in high-dimensional data stream.

Da-Zheng Feng1, Wei-Xing Zheng, Ying Jia

  • 1National Laboratory for Radar Signal Processing, Xidian University, 710071 Xi'an, PR China. dzfeng@xidian.edu.cn

IEEE Transactions on Neural Networks
|June 9, 2005
PubMed
Summary
This summary is machine-generated.

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

Chrysin reprograms microglial metabolism and function via targeting SYK to alleviate the symptoms of Alzheimer's disease.

Journal of ethnopharmacology·2026
Same author

High-energy-density aqueous magnesium metal battery textiles enable ultrasensitive pressure sensing across -40° to 100°C.

Science advances·2026
Same author

Research on Coal Gangue Image Recognition Method Based on EMAM-YOLO.

ACS omega·2026
Same author

Emergency decontamination of leaked unsymmetrical dimethylhydrazine with carboxyl-rich graphene oxide: performance and mechanism.

RSC advances·2026
Same author

Immunomodulatory alginate microcapsules for enhanced delivery of purple sweet potato anthocyanins.

Food science and biotechnology·2026
Same author

SOX9, the Master Regulator of Lung Cancer, and a Therapeutic Approach.

Journal of biochemical and molecular toxicology·2026
Same journal

Universal perceptron and DNA-like learning algorithm for binary neural networks: LSBF and PBF implementations.

IEEE transactions on neural networks·2013
Same journal

Data-based fault-tolerant control of high-speed trains with traction/braking notch nonlinearities and actuator failures.

IEEE transactions on neural networks·2011
Same journal

Neural network-based multiple robot simultaneous localization and mapping.

IEEE transactions on neural networks·2011
Same journal

Data-driven model-free adaptive control for a class of MIMO nonlinear discrete-time systems.

IEEE transactions on neural networks·2011
Same journal

Data-based identification and control of nonlinear systems via piecewise affine approximation.

IEEE transactions on neural networks·2011
Same journal

Data-core-based fuzzy min-max neural network for pattern classification.

IEEE transactions on neural networks·2011
See all related articles

A new algorithm, OJAm, efficiently tracks the minor subspace (MS) in data streams. It overcomes divergence issues and uniquely identifies an orthonormal basis for the MS, improving upon existing methods.

Area of Science:

  • Signal Processing
  • Machine Learning
  • Linear Algebra

Background:

  • Online tracking of the minor component (MC) and minor subspace (MS) is crucial for analyzing vector data streams.
  • Existing random-gradient algorithms for MC tracking face dynamical divergence issues.

Purpose of the Study:

  • To develop a novel algorithm for online tracking of the minor subspace (MS) of autocorrelation matrices.
  • To extend existing single MC tracking algorithms to handle multiple MCs or the MS.
  • To address and overcome the dynamical divergence limitations of current algorithms.

Main Methods:

  • A modified Oja-type algorithm, termed OJAm, is proposed.
  • Analysis of the averaging differential equation and energy (Lyapunov) functions associated with OJAm.

Related Experiment Videos

  • Global convergence properties of OJAm are investigated.

    • Main Results:

    • OJAm demonstrates satisfactory performance and overcomes dynamical divergence.
    • The averaging differential equation for OJAm globally asymptotically converges to an invariance set.
    • OJAm uniquely identifies a global minimum corresponding to the MS, unlike other algorithms.
    • OJAm can track an orthonormal basis of the MS, a capability lacking in previous methods.

    Conclusions:

    • OJAm offers an efficient online learning approach for tracking the MS.
    • The algorithm provides a robust and accurate method for subspace tracking in data streams.
    • OJAm significantly advances the state-of-the-art in online minor subspace analysis.