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 generalized feedforward neural network architecture for classification and regression.

Ganesh Arulampalam1, Abdesselam Bouzerdoum

  • 1Edith Cowan University, 100 Joondalup Drive, Joondalup, WA 6027, Australia. g.arulampalam@ecu.edu.au

Neural Networks : the Official Journal of the International Neural Network Society
|July 10, 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

Correcting Susceptibility Artifacts of MRI Sensors in Brain Scanning: A 3D Anatomy-Guided Deep Learning Approach.

Sensors (Basel, Switzerland)·2021
Same author

Unbalanced Hybrid AOA/RSSI Localization for Simplified Wireless Sensor Networks.

Sensors (Basel, Switzerland)·2020
Same author

An unsupervised deep learning technique for susceptibility artifact correction in reversed phase-encoding EPI images.

Magnetic resonance imaging·2020
Same author

Compressive Radar Imaging of Stationary Indoor Targets with Low-rank plus Jointly Sparse and Total Variation Regularizations.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2020
Same author

Hybrid Deep Learning-Gaussian Process Network for Pedestrian Lane Detection in Unstructured Scenes.

IEEE transactions on neural networks and learning systems·2020
Same author

DeepDSSR: Deep Learning Structure for Human Donor Splice Sites Recognition.

Studies in health technology and informatics·2019

A new generalized feedforward neural network (GFNN) architecture, utilizing generalized shunting neurons (GSNs), demonstrates superior performance in pattern classification. This novel GFNN approach surpasses traditional methods like SIANNs and multilayer perceptrons in benchmark tests.

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Computational Neuroscience

Background:

  • Traditional neural networks like perceptrons and shunting inhibitory artificial neural networks (SIANNs) have limitations in forming complex decision boundaries.
  • Existing architectures may struggle with intricate pattern classification and regression tasks.

Purpose of the Study:

  • Introduce a novel generalized feedforward neural network (GFNN) architecture.
  • Explore the capabilities of the generalized shunting neuron (GSN) as a fundamental computing unit.
  • Evaluate the GFNN architecture's effectiveness in pattern classification and regression.

Main Methods:

  • Developed a new GFNN architecture employing generalized shunting neurons (GSNs).
  • GSNs integrate perceptron and shunting inhibitory neuron models, enabling nonlinear decision boundaries.

Related Experiment Videos

  • Applied GFNNs to several benchmark classification problems for performance evaluation.

    • Main Results:

    • GFNNs, powered by GSNs, effectively learn complex, nonlinear decision boundaries.
    • Experimental results indicate that a single GSN can outperform both SIANN and multilayer perceptron (MLP) networks.
    • The GFNN architecture shows significant potential for complex pattern classification tasks.

    Conclusions:

    • The proposed GFNN architecture offers a powerful new approach for pattern classification and regression.
    • GSNs provide a versatile computational unit capable of handling complex data patterns.
    • GFNNs represent an advancement over existing neural network models, particularly in classification accuracy.