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

Development of low entropy coding in a recurrent network.

G F Harpur1, R W Prager

  • 1Engineering Department, University of Cambridge, Trumpington Street, Cambridge, CB2 1PZ, UK.

Network (Bristol, England)
|May 1, 1996
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

Three-dimensional ultrasound imaging.

Proceedings of the Institution of Mechanical Engineers. Part H, Journal of engineering in medicine·2010
Same author

Initial clinical experience of an ultrasonic strain imaging system with novel noise-masking capability.

The British journal of radiology·2010
Same author

3-D ultrasonic strain imaging using freehand scanning and a mechanically-swept probe.

IEEE transactions on ultrasonics, ferroelectrics, and frequency control·2010
Same author

Evaluation of a three-dimensional ultrasound localisation system incorporating probe pressure correction for use in partial breast irradiation.

The British journal of radiology·2009
Same author

Spatial mapping of the brachial plexus using three-dimensional ultrasound.

The British journal of radiology·2005
Same author

3D reconstruction of the skeletal anatomy of the normal neonatal foot using 3D ultrasound.

The British journal of radiology·2005
Same journal

Enhancing IoT security: A Creative Swagger Optimization algorithm for DDoS defence.

Network (Bristol, England)·2026
Same journal

Parametric optimization for electrical discharge diamond grinding (EDDG) system using dual approach.

Network (Bristol, England)·2025
Same journal

A novel lung cancer diagnosis model using hybrid convolution (2D/3D)-based adaptive DenseUnet with attention mechanism.

Network (Bristol, England)·2025
Same journal

Hybrid optimization enabled Eff-FDMNet for Parkinson's disease detection and classification in federated learning.

Network (Bristol, England)·2025
Same journal

AI-driven plant disease detection with tailored convolutional neural network.

Network (Bristol, England)·2025
Same journal

Layer modified residual Unet++ for speech enhancement using Aquila Black widow optimizer algorithm.

Network (Bristol, England)·2025
See all related articles

This study introduces an unsupervised neural network that minimizes reconstruction error using competitive units. It achieves non-orthogonal weights, differing from principal components analysis, for enhanced information transfer and reduced code entropy.

Area of Science:

  • Computational Neuroscience
  • Machine Learning
  • Information Theory

Background:

  • Unsupervised neural networks are crucial for pattern recognition and data compression.
  • Existing methods like principal components analysis often yield orthogonal solutions.
  • Minimizing reconstruction error is a key objective in unsupervised learning.

Purpose of the Study:

  • To present a novel unsupervised neural network architecture.
  • To demonstrate its ability to converge to non-orthogonal weight values.
  • To explore its application in information transfer and entropy reduction.

Main Methods:

  • Developed an unsupervised neural network with inhibitory feedback between units.
  • The network is trained to minimize reconstruction error for individual patterns and the entire dataset.

Related Experiment Videos

  • Explored the relationship between network operation, information transfer, and code entropy.

    • Main Results:

    • The network successfully minimizes reconstruction error.
    • It demonstrates convergence to non-orthogonal weight values, a key distinction from PCA.
    • Results show the effectiveness of assigning prior probabilities to reduce code entropy.

    Conclusions:

    • The proposed unsupervised neural network offers a novel approach to data representation.
    • Its ability to achieve non-orthogonal weights provides advantages over traditional methods.
    • The network shows promise for applications in binary coding and image compression.