A Modified General Regression Neural Network (MGRNN) with new, efficient training algorithms as a robust 'black box'-tool for data analysis

Authors:
Dirk Tomandl;Andreas Schober
Affiliations:
Institute for Physical High Technology Jena (IPHT), Micro Systems Division, Postfach 10 02 39, 07702 Jena, Germany and Fa. Merck KGaA, Abt Biomed-FO HK (A31/105), NanoSynTest, 64271 Darmstadt, Ger ...;Institute for Physical High Technology Jena (IPHT), Micro Systems Division, Postfach 10 02 39, 07702 Jena, Germany and Fa. Merck KGaA, Abt Biomed-FO HK (A31/105), NanoSynTest, 64271 Darmstadt, Ger ...
Venue:
Neural Networks
Year:
2001

Citing 5
Cited 13

Original Contribution: Mapping neural network derived from the parzen window estimator

Neural Networks
Numerical recipes in C (2nd ed.): the art of scientific computing

Numerical recipes in C (2nd ed.): the art of scientific computing
The general regression neural network-Rediscovered

Neural Networks
Advanced algorithms for neural networks: a C++ sourcebook

Advanced algorithms for neural networks: a C++ sourcebook
Learning experiments with genetic optimization of a generalized regression neural network

Decision Support Systems - Special double issue: unified programming

Evolutionary product unit based neural networks for regression

Neural Networks
Variations of the two-spiral task

Connection Science
Appearance-Based Map Learning for Mobile Robot by Using Generalized Regression Neural Network

ISNN '07 Proceedings of the 4th international symposium on Neural Networks: Advances in Neural Networks
Kernel regression networks with local structural information and covariance volume adaptation

Neurocomputing
Feature subset selection in large dimensionality domains

Pattern Recognition
Application of a noisy data classification technique to determine the occurrence of flashover in compartment fires

Advanced Engineering Informatics
Incremental learning of spatio-temporal patterns with model selection

ICANN'07 Proceedings of the 17th international conference on Artificial neural networks
Research on the electric load forecasting and risk assessment based on wavelet neural network

IITA'09 Proceedings of the 3rd international conference on Intelligent information technology application
A neural network-based framework for the reconstruction of incomplete data sets

Neurocomputing
Comparative study on input-expansion-based improved general regression neural network and levenberg-marquardt BP network

ICIC'06 Proceedings of the 2006 international conference on Intelligent Computing - Volume Part I
Determinants Of Capital Structure In The Uk Retail Industry: A Comparison Of Multiple Regression And Generalized Regression Neural Network

International Journal of Intelligent Systems in Accounting and Finance Management
Generalized classifier neural network

Neural Networks
Artificial neural network analysis of sierpinski gasket fractal antenna: a low cost alternative to experimentation

Advances in Artificial Neural Systems

Quantified Score

Hi-index	0.01

Visualization

Abstract

A Modified General Regression Neural Network (MGRNN) is presented as an easy-to-use 'black box'-tool to feed in available data and obtain a reasonable regression surface. The MGRNN is based on the General Regression Neural Network by D. Specht [Specht, D. (1991). A General Regression Neural Network. IEEE Transactions on Neural Networks, 2(6), 568-576], therefore, the network's architecture and weights are determined. The kernel width of each training sample is trained by two supervised training algorithms. These fast and reliable algorithms require four user-definable parameters, but are robust against changes of the parameters. Its generalization ability was tested with different benchmarks: intertwined spirals, Mackey--Glass time series and PROBEN1. The MGRNN provides two additional features: (1) it is trainable with arbitrary data as long as a suitable metric exists. Particularly, it is unnecessary to force the data structure to vectors of equal length; (2) it is able to compute the gradient of the regression surface as long as the gradient of the metric is definable and defined. The MGRNN solves common practical problems of common feed-forward networks.