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 new neural network for solving nonlinear projection equations.

Youshen Xia1, Gang Feng

  • 1College of Mathematics and Computer Science, Fuzhou University, Fuzhou 350002, China. ysxia2001@yahoo.com

Neural Networks : the Official Journal of the International Neural Network Society
|April 25, 2007
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

The neural correlates of value hierarchies: a prospective typology based on personal value profiles of emerging adults.

Frontiers in psychology·2024
Same author

Efficacy and Safety of PF-07038124 in Patients With Atopic Dermatitis and Plaque Psoriasis: A Randomized Clinical Trial.

JAMA dermatology·2023
Same author

Improving enteral nutrition tolerance and protein intake maybe beneficial to intensive care unit patients.

Scientific reports·2023
Same author

Efficacy and Safety of Dual Anti-HER2 Blockade and Docetaxel With or Without Carboplatin as Neoadjuvant Regimen for Treatment of HER2-Positive Breast Cancer.

Technology in cancer research & treatment·2023
Same author

Boosting Low-Temperature CO<sub>2</sub> Hydrogenation over Ni-based Catalysts by Tuning Strong Metal-Support Interactions.

Angewandte Chemie (International ed. in English)·2023
Same author

A new direction in Chinese herbal medicine ameliorates for type 2 diabetes mellitus: Focus on the potential of mitochondrial respiratory chain complexes.

Journal of ethnopharmacology·2023
Same journal

Anchor-based disentanglement framework for incremental multi-view clustering.

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

Complex-valued amplitude-phase interference modeling for adversarially robust classification.

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

TraNce: Type-aware hypergraph neural network with biological mediators for drug repositioning.

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

Decentralized ADMM for factorization-based Low-rank matrix estimation.

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

Memristive neuromorphic circuit design inspired by the neural mechanisms of conditioned fear.

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

A novel recurrent neural network efficiently solves nonlinear projection equations. This new model offers global convergence and enhanced stability for broader optimization problems, outperforming existing methods.

Area of Science:

  • Artificial Intelligence
  • Computational Neuroscience
  • Optimization Theory

Background:

  • Nonlinear projection equations are fundamental in various fields, including optimization and machine learning.
  • Existing recurrent neural networks for solving these equations often require smooth nonlinear mappings and lack guaranteed stability.
  • There is a need for robust neural network architectures that can handle non-smooth mappings and ensure convergence.

Purpose of the Study:

  • To propose a new one-layer recurrent neural network for solving nonlinear projection equations.
  • To demonstrate the network's global convergence and stability properties under mild conditions.
  • To establish the network's applicability to a wider range of optimization and related problems.

Main Methods:

Related Experiment Videos

  • Development of a novel one-layer recurrent neural network architecture.
  • Theoretical analysis to prove global convergence and stability (asymptotic and exponential) without requiring smooth nonlinear mappings.
  • Comparative analysis with existing neural network methods using illustrative examples.

    • Main Results:

    • The proposed neural network guarantees global convergence to an exact solution under mild conditions.
    • It achieves asymptotic and exponential stability even for non-smooth nonlinear mappings.
    • The network effectively finds equilibrium points for projection and Hopfield-type neural networks.

    Conclusions:

    • The new recurrent neural network is a versatile and effective solver for a broad class of nonlinear projection and optimization problems.
    • It offers improved accuracy and faster convergence rates compared to existing methods.
    • Its parallel implementation suitability and robustness make it a valuable tool in computational science.