Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

Coupled principal component analysis.

Ralf Möller, Axel Könies

IEEE Transactions on Neural Networks

|September 25, 2004

Summary

This summary is machine-generated.

This study introduces coupled principal component learning rules that simultaneously estimate eigenvectors and eigenvalues. These new rules overcome stability-speed issues in traditional methods for faster, more reliable principal component analysis.

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

H-NGPCA: Hierarchical clustering of data streams with adaptive number of clusters and adaptive dimensionality.

PloS one·2026

Same author

Excitation of Alfvénic Modes via Electromagnetic Turbulence in Wendelstein 7-X.

Physical review letters·2025

Same author

Earth's most needed uncultivated aquatic prokaryotes.

Water research·2024

Same author

Data integration and analysis for circadian medicine.

Acta physiologica (Oxford, England)·2023

Same author

Adaptive dimensionality reduction for neural network-based online principal component analysis.

PloS one·2021

Same author

Emergency ventilator for COVID-19.

PloS one·2020

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

Guest editorial: special section on white box nonlinear prediction models.

IEEE transactions on neural networks·2011

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

Guest editorial: special section on data-based control, modeling, and optimization.

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

See all related articles

Area of Science:

Machine Learning
Linear Algebra
Signal Processing

Background:

Traditional principal component analysis (PCA) learning rules face a stability-speed tradeoff, limiting their efficiency in dynamic systems.
Existing noncoupled rules exhibit convergence speeds dependent on covariance matrix eigenvalues, causing instability.

Discussion:

Presents a novel framework for coupled principal component learning rules, estimating eigenvectors and eigenvalues simultaneously.
Derived new coupled learning rule systems for PCA using Newton's method on an information criterion.
Establishes relationships between the proposed coupled rules and existing methods like ALA and RRLSA.

Key Insights:

Coupled rules mitigate the stability-speed problem by decoupling convergence speed from covariance matrix eigenvalues.

Related Experiment Videos

Simultaneous estimation in coupled equations leads to improved performance across all eigendirections.

Two new coupled learning rule systems for PCA are introduced, enhancing the existing class of algorithms.

Outlook:

Potential for application in adaptive filtering, dimensionality reduction, and feature extraction.
Further research can explore extensions to non-linear PCA and other matrix decomposition techniques.
Investigating the theoretical convergence properties and practical performance across diverse datasets is warranted.