Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

Efficient training of RBF networks for classification.

¹Neural Computing Research Group, Aston University, Birmingham, B4 7ET, UK. i.t.nabney@aston.ac.uk

International Journal of Neural Systems

|July 10, 2004

Summary

Radial Basis Function (RBF) networks can be efficiently trained for classification using logistic and softmax outputs with the Fisher scoring algorithm, offering a faster alternative to standard optimization methods.

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

Remote Patient Monitoring and Machine Learning in Acute Exacerbations of Chronic Obstructive Pulmonary Disease: Dual Systematic Literature Review and Narrative Synthesis.

Journal of medical Internet research·2024

Same author

Exacerbation predictive modelling using real-world data from the myCOPD app.

Heliyon·2024

Same author

Detecting dynamical changes in vital signs using switching Kalman filter.

Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2017

Same author

Scrutinizing human MHC polymorphism: Supertype analysis using Poisson-Boltzmann electrostatics and clustering.

Journal of molecular graphics & modelling·2017

Same author

Early warnings of heart rate deterioration.

Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2017

Same author

Regularity analysis of nocturnal oximetry recordings to assist in the diagnosis of sleep apnoea syndrome.

Medical engineering & physics·2016

Area of Science:

Machine Learning
Artificial Intelligence
Computational Statistics

Background:

Radial Basis Function (RBF) networks with linear outputs are commonly used for regression due to their training speed compared to Multi-layer Perceptrons.
Linear outputs in RBF networks are unsuitable for classification as they do not ensure probability representation.

Purpose of the Study:

To demonstrate an efficient training method for RBF networks with logistic and softmax outputs for classification tasks.
To introduce the Fisher scoring algorithm as a viable training approach for generalized linear models.

Main Methods:

Implementing the Fisher scoring algorithm for training RBF networks with logistic and softmax output functions.
Applying the method to models with a generalized linear output function on top of a linear model.
Comparing the Fisher scoring approach against standard non-linear optimization algorithms.

Main Results:

The Fisher scoring algorithm enables efficient training of RBF networks with logistic and softmax outputs.
This method is applicable to a broad class of models linear in their parameters with generalized linear outputs.
Performance comparisons on various datasets indicate the effectiveness of the proposed approach.

Conclusions:

The Fisher scoring algorithm provides an efficient training solution for RBF networks in classification.
This generalized approach enhances the applicability of RBF networks beyond regression problems.
The method offers a competitive alternative to traditional non-linear optimization techniques for specific model architectures.

Related Experiment Videos