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

Features selection and architecture optimization in connectionist systems.

M Yacoub1, Y Bennani

  • 1LIPN-CNRS, Institut Galilée Université Paris 13, Villetaneuse, France. yacoub@lipn.univ-paris13.fr

International Journal of Neural Systems
|February 24, 2001
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

Risk factors for RNA integrity in fresh-frozen renal cancer samples from routine practice (UroCCR 9).

The French journal of urology·2026
Same author

Segmental 3D image-guided robot-assisted partial nephrectomy (3D-IGRAPN) in selected cases of localized renal urothelial carcinoma.

World journal of urology·2025
Same author

Testotoxicosis without Testicular Mass: Revealed by Peripheral Precocious Puberty and Confirmed by Somatic <i>LHCGR</i> Gene Mutation.

Endocrine research·2019
Same author

The prognostic value of quantitative diffusion-weighted MRI after pediatric cardiopulmonary arrest.

Resuscitation·2018
Same author

Erratum to "In vivo percutaneous microwave ablation in kidneys: Correlation with ex vivo data and ablation work" [Diagn. Interv. Imaging 99 (1) (2018) 3-8].

Diagnostic and interventional imaging·2018
Same author

In vivo percutaneous microwave ablation in kidneys: Correlation with ex vivo data and ablation work.

Diagnostic and interventional imaging·2017
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

This study introduces HVS, a novel heuristic for variable selection, to optimize Multi-Layer Perceptron (MLP) architectures. The method effectively identifies crucial features, enhancing model accuracy in regression and discrimination tasks.

Area of Science:

  • Machine Learning
  • Artificial Intelligence
  • Neural Networks

Background:

  • Multi-Layer Perceptrons (MLPs) are fundamental in deep learning.
  • Feature selection and architecture optimization are critical for MLP performance.
  • Existing methods may not efficiently handle redundant or irrelevant features.

Purpose of the Study:

  • To introduce a new heuristic measure, HVS (Heuristic for Variable Selection), for feature selection.
  • To develop a procedure for optimizing MLP architecture using the HVS measure.
  • To extend the HVS approach (epsilonHVS) for Time Delay Neural Networks (TDNNs).

Main Methods:

  • The HVS heuristic identifies and selects important variables by eliminating redundant or uninformative features.
  • A novel procedure utilizes HVS to determine the optimal MLP architecture from an initial structure.

Related Experiment Videos

  • The epsilonHVS extension is proposed for discriminative feature detection and TDNN optimization.

    • Main Results:

    • The HVS-based algorithm effectively identifies optimized connectionist models.
    • Demonstrated effectiveness in both regression and discrimination tasks.
    • Achieved higher accuracy in identifying optimized MLP architectures.

    Conclusions:

    • The proposed HVS feature selection and MLP architecture optimization procedure is effective.
    • The epsilonHVS extension shows promise for TDNN models.
    • This approach enhances the accuracy and efficiency of neural network models.