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

Weight decay backpropagation for noisy data.

Amit Gupta1, Siuwa M. Lam

  • 1Andersen Consulting, USA

Neural Networks : the Official Journal of the International Neural Network Society
|March 29, 2003
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

Efficacy and safety of sofosbuvir-based antiviral therapy to treat hepatitis C virus infection after kidney transplantation.

Clinical kidney journal·2018
Same author

Solitary fibrous tumor of the liver - An unusual entity: A case report and review of literature.

Annals of hepato-biliary-pancreatic surgery·2018
Same author

Inadvertent Intrathecal Administration of Local Anesthetics Leading to Spinal Paralysis with Lipid Emulsion Rescue.

Clinical practice and cases in emergency medicine·2018
Same author

Management of CHAOS by intact cord resuscitation: case report and literature review.

The journal of maternal-fetal & neonatal medicine : the official journal of the European Association of Perinatal Medicine, the Federation of Asia and Oceania Perinatal Societies, the International Society of Perinatal Obstetricians·2018
Same author

A comprehensive review on Primary gallbladder tuberculosis.

Polski przeglad chirurgiczny·2018
Same author

Falciform Ligament Artery Uptake on 99mTc MAA Planning Scan Before 90Y SIRT Confirmed by Retrospective SPECT/MRI Fusion.

Clinical nuclear medicine·2018
Same journal

Q-learning based asynchronous Boolean control networks stabilization with data loss.

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

New results on prescribed-time synchronization of complex networks via intermittent control.

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

Variance-constrained multi-view ensemble broad network for imbalanced data.

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

Dynamic analysis and reliable mechanical optimization application of ring HNN effected with a memristive neuron.

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

DAFF-Net: A detection and search method for small-scale low surface brightness galaxies.

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

Quasi-synchronization for complex networks with hybrid pinning intermittent control.

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

Weight decay training enhances neural network robustness against noisy data. Keeping test data noise-free is crucial for accurate predictions, with weight decay outperforming standard backpropagation in most noisy scenarios.

Area of Science:

  • Machine Learning
  • Artificial Intelligence
  • Neural Networks

Background:

  • Weight decay is proposed as a method to improve neural network robustness against noisy datasets.
  • Investigating neural network performance under various noise conditions is essential for practical applications.

Purpose of the Study:

  • To formally evaluate the impact of weight decay training on backpropagation with noisy datasets.
  • To compare the effectiveness of weight decay versus standard backpropagation across different noise scenarios.
  • To determine the optimal training duration for weight decay and standard backpropagation in noisy environments.

Main Methods:

  • Formal evaluation of weight decay training for backpropagation.
  • Experimentation across three noisy data scenarios: noisy training/clean testing, clean training/noisy testing, and noisy training/noisy testing.

Related Experiment Videos

  • Comparison of performance metrics between weight decay and standard backpropagation.

    • Main Results:

    • The scenario with a noisy training set and a clean test set yielded more accurate predictions.
    • Weight decay training performed comparably to or better than standard backpropagation in noisy situations.
    • For clean datasets, no significant performance difference was observed between the two training methods.
    • Weight decay achieved faster convergence, outperforming standard backpropagation with shorter training epochs.
    • Extended training improved standard backpropagation but degraded weight decay performance.

    Conclusions:

    • Maintaining noise-free test data is critical for accurate neural network classification.
    • Weight decay training offers a robust alternative to standard backpropagation, especially in noisy environments.
    • The number of training epochs significantly influences the comparative performance of weight decay and standard backpropagation on noisy data.