Title :
On model selection in SLT and linear basis neural networks
Author :
Aguirre, Arturo Hernández ; Koutsougeras, Cris ; Buckles, Bill P.
Author_Institution :
Dept. of Electr. Eng. & Comput. Sci., Tulane Univ., New Orleans, LA, USA
Abstract :
This paper presents an approach to the experimental verification of the quality of “model selection” delivered by statistical learning theory (SLT). We depart from a function whose analytical approximation properties by polynomials are well known and readily verifiables in our experimental environment. For different sample size sets, the model predicted by SLT is contrasted against the model derived from the mathematical properties of the function. We found great robustness in the predictive ability of SLT
Keywords :
function approximation; learning (artificial intelligence); least squares approximations; neural nets; polynomials; statistical analysis; analytical approximation; least squares approximation; linear basis neural networks; model selection; polynomials; statistical learning theory; Approximation error; Chebyshev approximation; Function approximation; Intelligent networks; Machine learning; Mathematical model; Neural networks; Polynomials; Predictive models; Statistical learning;
Conference_Titel :
Neural Networks, 1999. IJCNN '99. International Joint Conference on
Conference_Location :
Washington, DC
Print_ISBN :
0-7803-5529-6
DOI :
10.1109/IJCNN.1999.831540
