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 local training and pruning approach for neural networks.

S J Chang1, C S Leung, K W Wong

  • 1Department of Electronic Engineering, City University of Hong Kong, Kowloon Tong.

International Journal of Neural Systems
|April 20, 2001
PubMed
Summary

We introduce a local extended Kalman filter (EKF) training and pruning method to reduce computational demands for neural network training. This approach significantly lowers complexity and storage needs, making EKF training more practical.

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

MRI evaluation of congenital vaginal malformations-a pictorial review.

Clinical radiology·2026
Same author

The incidence of arthrogenic muscle inhibition in acute knee injuries: a prospective cohort study in a tertiary acute knee unit.

Asia-Pacific journal of sports medicine, arthroscopy, rehabilitation and technology·2025
Same author

Direct Experimental Proof of the Principal Role of Reduced High-Mode Hydrodynamic Mix in Recent Ignition Success on NIF.

Physical review letters·2025
Same author

Exploring global barriers to optimal ovarian cancer care: thematic analysis.

International journal of gynecological cancer : official journal of the International Gynecological Cancer Society·2025
Same author

Physiological and molecular impairment of PV circuit homeostasis in mouse models of autism.

bioRxiv : the preprint server for biology·2025
Same author

Time Pattern of Presentation of Victims of High-Speed Passenger Ferry Mass Casualty Incidents to the Emergency Department.

Disaster medicine and public health preparedness·2024

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Computational Neuroscience

Background:

  • Training neural networks with the extended Kalman filter (EKF) algorithm presents significant computational and storage challenges.
  • These challenges can limit the applicability of EKF for even moderately sized neural networks.

Purpose of the Study:

  • To present a novel local EKF training and pruning approach to overcome the computational drawbacks of traditional EKF training.
  • To demonstrate a more practical and efficient method for training neural networks using EKF.

Main Methods:

  • Developed a local EKF training methodology.
  • Utilized by-products from local EKF training to assess the importance of network weights for pruning.
  • Compared the proposed local approach against the original global EKF training method.

Related Experiment Videos

Main Results:

  • The local EKF training and pruning approach significantly reduces computational complexity and storage requirements compared to the global method.
  • The method proved effective in addressing real-world problems, demonstrating its practical utility.
  • Successful performance was shown on medium- and large-scale tasks, including sunspot prediction and handwritten digit recognition.

Conclusions:

  • The proposed local EKF training and pruning method offers a computationally efficient and practical alternative for training neural networks.
  • This approach effectively addresses the limitations of traditional EKF, enabling its use in more complex applications.
  • The demonstrated success on diverse datasets highlights the algorithm's robustness and scalability.