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

Spatio-temporal feature maps using gated neuronal architecture.

V Chandrasekaran1, M Palaniswami, T M Caelli

  • 1Dept. of Electr. and Electron. Eng., Melbourne Univ., Parkville, Vic.

IEEE Transactions on Neural Networks
|January 1, 1995
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

Fungal infection in a post-renal transplant patient with <i>Diaporthe</i>.

Indian journal of medical microbiology·2020
Same author

Residue from vacuum ultraviolet irradiation of benzene ices: Insights into the physical structure of astrophysical dust.

Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy·2019
Same author

Intrapatient variability in plasma rifampicin & isoniazid in tuberculosis patients.

The Indian journal of medical research·2018
Same author

Ultraslow isomerization in photoexcited gas-phase carbon cluster [Formula: see text].

Nature communications·2018
Same author

An experimental setup to study delayed electron emission upon photoexcitation of trapped polyatomic anions.

The Review of scientific instruments·2017
Same author

Assessing the development of fetal myocardial function by a novel Doppler myocardial performance index.

Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2017
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 a gated neural network architecture for Kohonen

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Computational Neuroscience

Background:

  • Kohonen's self-organizing feature map (SOFM) is a popular unsupervised learning algorithm.
  • Traditional SOFM lacks mechanisms for dynamic neuron competition, potentially limiting classification accuracy.
  • Feature space representation and neuron competition are key aspects of SOFM performance.

Purpose of the Study:

  • To enhance Kohonen's self-organizing feature map (SOFM) using a novel selective competition mechanism.
  • To improve classification performance by generating time sequences of winning node indexes.
  • To develop and evaluate new class label prediction algorithms based on evidential reasoning and Bayes' theorem.

Main Methods:

  • Modification of SOFM with a gated neuronal architecture enabling selective neuron competition.

Related Experiment Videos

  • Spatial frequency grating of the N-dimensional feature space to define competition criteria.
  • Generation of time sequences of winning node indexes based on changing selection criteria.
  • Formulation of three class label prediction algorithms using evidential reasoning and Bayes conditional probability.
  • Testing on real-world 8-class texture and synthetic 12-class 3D object recognition datasets.

    • Main Results:

    • The proposed gated SOFM architecture demonstrates improved classification performance.
    • Time sequences of winning node indexes provide richer input information for classification.
    • The developed prediction algorithms show competitive or superior results compared to standard linear discriminant analysis.
    • Effective classification achieved on both texture and 3D object recognition tasks.

    Conclusions:

    • The novel selective competition mechanism significantly enhances SOFM capabilities.
    • The time-varying winning node sequences offer a powerful feature for accurate pattern classification.
    • The proposed methods provide a promising approach for advanced pattern recognition and classification tasks.