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

A neural-network learning theory and a polynomial time RBF algorithm.

A Roy¹, S Govil, R Miranda

¹Dept. of Decision and Inf. Syst., Arizona State Univ., Tempe, AZ.

IEEE Transactions on Neural Networks

|January 1, 1997

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

Relation of family history of suicide to suicide attempts in alcoholics.

The American journal of psychiatry·2000

Same author

[Autoimmune hepatitis in children. Initial presentation as fulminant hepatic failure].

Acta gastroenterologica Latinoamericana·2000

Same author

Malnutrition and hypernatraemia in breastfed babies.

Annals of tropical paediatrics·2000

Same author

Somatosensory evoked potentials associated with thermal activation of type II Adelta mechanoheat nociceptive afferents.

The International journal of neuroscience·2000

Same author

Coronary artery bypass grafting without cardiopulmonary bypass in pheochromocytoma.

The Journal of thoracic and cardiovascular surgery·2000

Same author

Rate limitations of unitary event analysis.

Neural computation·2000

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 a novel brain-like learning theory and algorithm for neural networks. The new method enhances function approximation using a mixed radial basis function (RBF) net, offering robust and reliable learning.

Area of Science:

Artificial Intelligence
Computational Neuroscience
Machine Learning

Background:

Classical connectionist learning lacks brain-like computational characteristics.
Developing robust and reliable neural network algorithms requires adherence to specific learning principles.
Function approximation is a key challenge in neural network applications.

Purpose of the Study:

To introduce a new learning theory and algorithm for neural networks inspired by brain-like learning.
To develop an algorithm for generating a novel radial basis function (RBF) network for function approximation.
To demonstrate the algorithm's effectiveness on various complex problems.

Main Methods:

A new learning theory defining brain-like computational characteristics.

Related Experiment Videos

A novel algorithm for generating a "mixed" RBF net combining truncated RBF and other hidden units.

Utilizing random clustering and linear programming (LP) for net design and training.

Main Results:

The algorithm generates a "mixed" RBF net, deviating from typical RBF architectures.
Polynomial time complexity is proven for the algorithm.
Successful computational results demonstrated on Mackey-Glass chaotic time series, logistic map prediction, neuro-control, and time series forecasting problems.

Conclusions:

The proposed learning theory and algorithm offer a more brain-like approach to neural network learning.
The "mixed" RBF net generated by the algorithm shows strong performance in function approximation tasks.
The algorithm is versatile and can be implemented as an online adaptive algorithm.