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

Regularization with a Pruning Prior.

Lars Kai Hansen1, Cyril Goutte

  • 1Department of Mathematical Modelling, Technical University of Denmark, DK-2800 Lyngby, Denmark

Neural Networks : the Official Journal of the International Neural Network Society
|August 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

Challenges in explaining deep learning models for data with biological variation.

PloS one·2025
Same author

On convex decision regions in deep network representations.

Nature communications·2025
Same author

Image classification with symbolic hints using limited resources.

PloS one·2024
Same author

Using sequences of life-events to predict human lives.

Nature computational science·2024
Same author

Modulation transfer functions for audiovisual speech.

PLoS computational biology·2022
Same author

Uncovering Cortical Units of Processing From Multi-Layered Connectomes.

Frontiers in neuroscience·2022
Same journal

Exploiting audio-visual modalities in videos: Object detection via multi-stage bilateral coupling network.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Reliability-aware modality completion with cross-modal distillation for federated learning with missing modalities.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

IGFD-Net: Illumination-guided frequency decoupling for polarization image fusion.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Multiple-Strategies dung beetle optimizer and its applications in engineering optimization and bankruptcy prediction.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Aggregating global-scale pixel-wise forgery cues within a graph.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Finite-Time intermittent control for secure synchronization of Neutral-Type stochastic delayed neural networks under aperiodic DoS attacks.

Neural networks : the official journal of the International Neural Network Society·2026
See all related articles

This study explores a regularization prior and its pruning capabilities. Researchers compared its performance against traditional weight decay using Bayesian and generalization methods on a toy problem.

Area of Science:

  • Machine Learning
  • Statistical Modeling
  • Artificial Intelligence

Background:

  • Regularization is crucial for preventing overfitting in machine learning models.
  • Traditional methods like weight decay are widely used but may have limitations.
  • Understanding novel regularization techniques is essential for advancing model performance.

Purpose of the Study:

  • To investigate the pruning properties of a specific regularization prior.
  • To compare the behavior of this prior with traditional weight decay.
  • To evaluate the effectiveness of the regularization prior within Bayesian and generalization frameworks.

Main Methods:

  • A novel regularization prior was analyzed for its inherent pruning characteristics.
  • Bayesian inference and a generalization-based method were employed for analysis.

Related Experiment Videos

  • A simple toy problem was used to illustrate and test the regularization prior's behavior.

    • Main Results:

    • The regularization prior demonstrated notable pruning properties.
    • The prior's behavior was effectively illustrated through both Bayesian and generalization analyses.
    • Comparisons showed distinct behaviors between the novel prior and traditional weight decay.

    Conclusions:

    • The investigated regularization prior exhibits useful pruning capabilities.
    • The study provides insights into the application of this prior in machine learning.
    • Findings suggest potential advantages over traditional weight decay methods for specific tasks.