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

Function approximation using generalized adalines.

Jiann-Ming Wu1, Zheng-Han Lin, Pei-Hsun Hsu

  • 1Department of Applied Mathematics, National Dong Hwa University, Hualien 941, Taiwan. jmwu@mail.ndhu.edu.tw

IEEE Transactions on Neural Networks
|May 26, 2006
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

Comparison of transoral and bilateral axillo-breast approach endoscopic thyroidectomy: multicentre study.

BJS open·2026
Same author

Tackling similarities and differences in global practice guidelines for gastric cancer: a review on the latest Taiwan guidelines with Asia-Pacific, European and US guidelines.

Gastric cancer : official journal of the International Gastric Cancer Association and the Japanese Gastric Cancer Association·2026
Same author

Urinary adhesion molecules with ERK signaling as predictive biomarkers for kidney stones: a pilot study.

BMC urology·2026
Same author

Can Symptoms Predict Delayed Gastric Emptying in Diabetic Patients: A Multicenter Study to Revisit Gastroparesis.

Journal of neurogastroenterology and motility·2025
Same author

Red dichromatic imaging-guided peroral endoscopic myotomy: a safer approach to achalasia with concurrent esophageal varices (with video).

Gastrointestinal endoscopy·2025
Same author

Endoscopic sleeve gastroplasty combined with anti-obesity medication for better control of weight and diabetes.

Clinical endoscopy·2025
Same journal

Universal perceptron and DNA-like learning algorithm for binary neural networks: LSBF and PBF implementations.

IEEE transactions on neural networks·2013
Same journal

Guest editorial: special section on white box nonlinear prediction models.

IEEE transactions on neural networks·2011
Same journal

Data-based fault-tolerant control of high-speed trains with traction/braking notch nonlinearities and actuator failures.

IEEE transactions on neural networks·2011
Same journal

Guest editorial: special section on data-based control, modeling, and optimization.

IEEE transactions on neural networks·2011
Same journal

Neural network-based multiple robot simultaneous localization and mapping.

IEEE transactions on neural networks·2011
Same journal

Data-driven model-free adaptive control for a class of MIMO nonlinear discrete-time systems.

IEEE transactions on neural networks·2011
See all related articles

This study introduces generalized adalines (gadalines) for data-driven function approximation. This novel neural network approach outperforms existing methods like MLP and RBF for accurate function approximation tasks.

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Computational Neuroscience

Background:

  • Adaline networks are fundamental in neural computation.
  • Generalizing Adaline's threshold function is key to enhanced capabilities.
  • Data-driven function approximation remains a critical challenge in machine learning.

Purpose of the Study:

  • To propose a novel neural organization of generalized adalines (gadalines) for data-driven function approximation.
  • To introduce a generative component for stochastic target generation using gadaline activation.
  • To develop and evaluate a hybrid optimization method for gadaline network training.

Main Methods:

  • Generalization of Adaline threshold function to create K-state transfer functions.
  • Development of a generative component triggering independent normal variables.

Related Experiment Videos

  • Application of mixed integer and linear programming, solved by mean field annealing and gradient descent.
  • Implementation of a leave-one-out learning strategy for optimizing multiple generative components.

    • Main Results:

    • The proposed learning method optimizes parameters for a deterministic gadaline network.
    • Numerical simulations demonstrate successful function approximation with diverse datasets.
    • The gadaline network learning method shows superior performance compared to MLP and RBF methods.

    Conclusions:

    • The proposed neural organization of generalized adalines offers a powerful new tool for data-driven function approximation.
    • The hybrid optimization approach effectively trains gadaline networks for complex tasks.
    • This method presents a significant advancement over traditional neural network models in function approximation accuracy.