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

A constructive algorithm that converges for real-valued input patterns

N Burgess1

  • 1Department of Anatomy, University College, London, U.K.

International Journal of Neural Systems
|March 1, 1994
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

An IL6 variant -174 G-C alters cardiac hypertrophy but not cardiometabolic responses to HFD in mice.

npj metabolic health and disease·2026
Same author

Exercise has differential cardiometabolic effects in male and female mice on a high-fat diet.

Physiological reports·2026
Same author

Efficacy and safety of the EMS Swiss LithoClast® Trilogy for PCNL: results of the European multicentre prospective study on behalf of European Section of UroTechnology.

World journal of urology·2021
Same author

Differential effects of negative emotion on memory for items and associations, and their relationship to intrusive imagery.

Current opinion in behavioral sciences·2017
Same author

Neural mechanisms of self-location.

Current biology : CB·2014
Same author

Theta phase precession of grid and place cell firing in open environments.

Philosophical transactions of the Royal Society of London. Series B, Biological sciences·2013
Same journal

Latent Space Projections and Atlases, a Cautionary Tale in Deep Neuroimaging using Autoencoders.

International journal of neural systems·2026
Same journal

Transformer-Based Anomaly Detection for Neurodegenerative Screening in MRI Images.

International journal of neural systems·2026
Same journal

Discrete Wavelet Convolution for Learnable Time-Frequency Representation with Application to Seizure Prediction.

International journal of neural systems·2026
Same journal

Automatic Seizure Detection using Hierarchical Spectral-Temporal Feature Learning with an Imbalance-Aware Transformer.

International journal of neural systems·2026
Same journal

Pyramid Vision Transformer-Enhanced Conformer Network for Epileptic Seizure Recognition Using MultiChannel EEG Signals.

International journal of neural systems·2026
Same journal

A Time-Frequency Decoupled Contrastive Learning Framework for Electroencephalography-Based Parkinson's Disease Diagnosis.

International journal of neural systems·2026
See all related articles

A novel constructive algorithm integrates Cascade Correlation and Upstart for perceptron-like networks, proving convergence for pattern classification. This method efficiently builds hidden units for complex problems like N-bit parity.

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Neural Networks

Background:

  • Existing constructive algorithms face challenges with real-valued inputs.
  • The need for efficient hidden unit construction in neural networks is critical.
  • Hyper-spherical decision regions can simplify classification tasks.

Purpose of the Study:

  • To present a novel constructive algorithm combining Cascade Correlation and Upstart.
  • To prove convergence guarantees for classifying real-valued pattern vectors.
  • To demonstrate the algorithm's efficiency in constructing hidden units.

Main Methods:

  • A constructive algorithm integrating Cascade Correlation architecture and Upstart training.
  • Theoretical proof of convergence to zero errors for consistent classifications.

Related Experiment Videos

  • Modification of input patterns by adding an element to enable hyper-spherical decision regions.

    • Main Results:

    • Guaranteed convergence to zero errors for any consistent classification of real-valued patterns.
    • Demonstrated robust convergence in simulations.
    • Economical construction of hidden units observed in benchmark problems.

    Conclusions:

    • The presented algorithm offers a robust and efficient method for neural network construction.
    • The approach enhances existing constructive algorithms for real-valued inputs.
    • Successful application to challenging benchmark datasets like N-bit parity and twin spirals.