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

On self-organizing algorithms and networks for class-separability features.

C Chatterjee1, V P Roychowdhury

  • 1Newport Corp., Irvine, CA.

IEEE Transactions on Neural Networks
|January 1, 1997
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

Black Hole Spectroscopy and Tests of General Relativity with GW250114.

Physical review letters·2026
Same author

GW250114: Testing Hawking's Area Law and the Kerr Nature of Black Holes.

Physical review letters·2025
Same author

High-Statistics Measurement of Collins and Sivers Asymmetries for Transversely Polarized Deuterons.

Physical review letters·2024
Same author

Final COMPASS Results on the Transverse-Spin-Dependent Azimuthal Asymmetries in the Pion-Induced Drell-Yan Process.

Physical review letters·2024
Same author

Deciphering miRNA-lncRNA-mRNA interaction through experimental validation of miRNAs, lncRNAs, and miRNA targets on mRNAs in Cajanus cajan.

Plant biology (Stuttgart, Germany)·2024
Same author

Search for Subsolar-Mass Binaries in the First Half of Advanced LIGO's and Advanced Virgo's Third Observing Run.

Physical review letters·2022
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

This study introduces self-organizing learning algorithms and neural networks for effective feature extraction, enhancing class separability. The novel Q(-1/2) network and its applications in data analysis are detailed, demonstrating robust performance.

Area of Science:

  • Machine Learning
  • Artificial Intelligence
  • Computational Neuroscience

Background:

  • Class separability is crucial for effective pattern recognition.
  • Existing feature extraction methods may not optimally preserve class distinctions.
  • Self-organizing algorithms offer adaptive learning capabilities.

Purpose of the Study:

  • To develop self-organizing learning algorithms and neural networks for feature extraction.
  • To enhance the preservation of class separability in extracted features.
  • To introduce and validate the Q(-1/2) network and its applications.

Main Methods:

  • An adaptive algorithm for computing Q(-1/2) (correlation/covariance matrix) is presented.
  • Convergence of the Q(-1/2) algorithm is proven using stochastic approximation theory.

Related Experiment Videos

  • Feature extraction architectures are developed using the Q(-1/2) network, principal component analysis, and demonstrated for Gaussian data, LDA, and Bhattacharyya distance.

    • Main Results:

    • A single-layer linear network, the Q(-1/2) network, is described.
    • The Q(-1/2) network is integrated with principal component analysis for advanced feature extraction.
    • Two-layer network convergence is proven, and numerical studies show performance on multiclass random data.

    Conclusions:

    • The proposed self-organizing algorithms and neural networks effectively extract features that preserve class separability.
    • The Q(-1/2) network provides a foundational component for various feature extraction tasks.
    • The study validates the theoretical convergence and practical performance of the developed networks.