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

Reduced HyperBF networks: regularization by explicit complexity reduction and scaled Rprop-based training.

Rami N Mahdi¹, Eric Christian Rouchka

¹Department of Genetic Medicine, Weill Cornell, Medical College, New York, NY 10065, USA. raminmahdi@yahoo.com

IEEE Transactions on Neural Networks

|March 23, 2011

Summary

This summary is machine-generated.

Related Concept Videos

Reducing Line Loss

Reducing Line Loss

In a three-phase circuit, line loss is an indicator of energy dissipated as heat due to the resistance of transmission lines. To address this, incorporating transformers into the system—a step-up transformer at the source and a step-down transformer at the load—is a strategic solution. Two three-phase transformers are introduced to improve this.
With a step-up transformer at the source, the voltage is increased, thereby reducing the current in the transmission lines since power loss in...

Neural Regulation

Neural Regulation

Digestion begins with a cephalic phase that prepares the digestive system to receive food. When our brain processes visual or olfactory information about food, it triggers impulses in the cranial nerves innervating the salivary glands and stomach to prepare for food.

You might also read

Related Articles

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

Sort by

Same author

Exacerbation of hemorrhagic stroke and impaired hemostasis in mice deficient of the integrated stress response kinase heme-regulated inhibitor.

Brain circulation·2026

Same author

Gene Expression and Alternative Splicing Analysis in a Large-Scale Multiple Sclerosis Study.

International journal of molecular sciences·2024

Same author

Codon usage bias as a function of generation time and life expectancy.

Bioinformation·2012

Same author

Model Based Unsupervised Learning Guided by Abundant Background Samples.

Proceedings of the ... International Conference on Machine Learning. International Conference on Machine Learning·2011

Same author

Evidence of bias towards buffered codons in human transcripts.

Proceedings of the ... IEEE International Symposium on Signal Processing and Information Technology. IEEE International Symposium on Signal Processing and Information Technology·2011

Same author

RBF-TSS: identification of transcription start site in human using radial basis functions network and oligonucleotide positional frequencies.

PloS one·2009

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

A new regularization method for Hyper basis function (HyperBF) networks improves classification accuracy and reduces model complexity. The proposed training approach offers faster convergence for HyperBF network construction.

Area of Science:

Artificial Intelligence
Machine Learning
Neural Networks

Background:

Hyper basis function (HyperBF) networks are advanced neural networks with high learning capacity.
Their complexity can lead to overfitting and challenging simultaneous optimization of network parameters.
Existing training methods face computational challenges with large datasets and network structures.

Purpose of the Study:

To introduce a novel regularization technique for HyperBF networks.
To develop a practical and efficient training methodology for constructing HyperBF networks.
To enhance the generalization performance and reduce the structural complexity of HyperBF networks.

Main Methods:

A new regularization method combining soft local dimension reduction and weight decay was proposed.

Related Experiment Videos

A hierarchical clustering approach was used for neuron initialization.

A modified Rprop algorithm with localized partial backtracking was employed for gradient optimization.

Main Results:

The regularized HyperBF network achieved classification accuracy comparable to support vector machines.
A significantly smaller network structure was required for the regularized HyperBF network.
The proposed training method demonstrated faster and smoother convergence compared to the standard Rprop algorithm across seven datasets.

Conclusions:

The proposed regularization and training method effectively addresses overfitting and computational challenges in HyperBF networks.
This approach enables the construction of more efficient and accurate HyperBF networks.
The findings suggest a practical and superior alternative for training generalized radial basis function networks.