Menu
Home
People
Places
Arts
History
Plants & Animals
Science
Life & Culture
Technology
Reference.org
General regression neural network
open-in-new
<p> Generalized regression neural network (GRNN) is a variation to <a href="/facts/Radial_basis_function_network/KBesEWD3">radial basis</a> <a href="/facts/Neural_network/DDhJTfMc">neural networks</a>. GRNN was suggested by D.F. Specht in 1991. </p><p>GRNN can be used for <a href="/facts/Regression_analysis/n6z5Tf7K">regression</a>, <a href="/facts/Prediction/0w4XXB33">prediction</a>, and <a href="/facts/Statistical_classification/jXXHRkXR">classification</a>. GRNN can also be a good solution for online <a href="/facts/Dynamical_system/782gREl5">dynamical systems</a>. </p><p>GRNN represents an improved technique in the neural networks based on the <a href="/facts/Nonparametric_regression/3MuUSUFV">nonparametric regression</a>. The idea is that every training sample will represent a mean to a radial basis <a href="/facts/Neuron/G7CyTVdH">neuron</a>. </p>