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

Performance evaluation of a sequential minimal radial basis function (RBF) neural network learning algorithm.

Y Lu1, N Sundararajan, P Saratchandran

  • 1School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore.

IEEE Transactions on Neural Networks
|February 7, 2008
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

Mechanistic Models: Compartment Models in Algorithms for Numerical Problem Solving01:29

Mechanistic Models: Compartment Models in Algorithms for Numerical Problem Solving

Mechanistic models play a crucial role in algorithms for numerical problem-solving, particularly in nonlinear mixed effects modeling (NMEM). These models aim to minimize specific objective functions by evaluating various parameter estimates, leading to the development of systematic algorithms. In some cases, linearization techniques approximate the model using linear equations.
In individual population analyses, different algorithms are employed, such as Cauchy's method, which uses a...
Response Surface Methodology01:16

Response Surface Methodology

Response Surface Methodology (RSM) is a collection of statistical and mathematical techniques used to develop, improve, and optimize processes. It is particularly valuable when many input variables or factors potentially influence a response variable.
The process of RSM involves several key steps:

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Identification of brain regions responsible for Alzheimer's disease using a Self-adaptive Resource Allocation Network.

Neural networks : the official journal of the International Neural Network Society·2012
Same author

A meta-cognitive learning algorithm for a Fully Complex-valued Relaxation Network.

Neural networks : the official journal of the International Neural Network Society·2012
Same author

Metacognitive learning in a fully complex-valued radial basis function neural network.

Neural computation·2011
Same author

Comparison of sensory properties of hamburgers cooked by conventional and carcinogen reducing `safe grill' equipment.

Meat science·2011
Same author

A fully complex-valued radial basis function network and its learning algorithm.

International journal of neural systems·2009
Same author

Online sequential fuzzy extreme learning machine for function approximation and classification problems.

IEEE transactions on systems, man, and cybernetics. Part B, Cybernetics : a publication of the IEEE Systems, Man, and Cybernetics Society·2009

The minimal resource allocation network (M-RAN) algorithm creates smaller neural networks with comparable or better accuracy than traditional methods. This efficient learning algorithm significantly reduces training time by avoiding data repetition.

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Computational Neuroscience

Background:

  • Radial basis function networks (RBFNs) are powerful function approximators.
  • Resource Allocating Networks (RANs) offer sequential learning but can grow excessively.
  • Efficient neural network pruning is crucial for reducing computational complexity.

Purpose of the Study:

  • To introduce and analyze the performance of the Minimal Resource Allocating Network (M-RAN) algorithm.
  • To compare M-RAN's efficiency and accuracy against established neural network training algorithms.
  • To demonstrate M-RAN's capability in achieving minimal network topologies.

Main Methods:

  • M-RAN combines RAN's growth criterion with a hidden unit contribution-based pruning strategy.

Related Experiment Videos

  • Performance evaluation involved benchmark function approximation and pattern classification tasks.
  • Comparative analysis against Multilayer Feedforward Networks (MFNs) trained with RPROP and Dependence Identification (DI) algorithms.

    • Main Results:

    • M-RAN consistently generated networks with significantly fewer hidden neurons compared to MFNs.
    • Achieved equal or superior approximation and classification accuracy.
    • Demonstrated considerably shorter learning (training) times due to single-pass data presentation.

    Conclusions:

    • M-RAN offers a more efficient approach to building RBFNs, achieving minimal topology.
    • The algorithm provides a favorable trade-off between network size, accuracy, and training speed.
    • M-RAN presents a promising alternative for resource-constrained machine learning applications.