Title :
On approximations of functions by depth-two neural networks
Author :
Venkatesh, Santosh S.
Author_Institution :
Dept. of Electr. Eng., Pennsylvania Univ., Philadelphia, PA, USA
fDate :
27 Jun-1 Jul 1994
Abstract :
The simple Pythagorean notion of orthogonal projections is used to show that depth-two sigmoidal neural networks can approximate any square-integrable function with compact support in Rn with arbitrarily small integrated squared-error
Keywords :
approximation theory; error analysis; function approximation; neural nets; Pythagorean notion; approximations; depth-two sigmoidal neural networks; integrated squared-error; orthogonal projections; square-integrable function; Convergence; Frequency locked loops; Function approximation; Hilbert space; Hypercubes; Multi-layer neural network; Neural networks; Neurons; Terminology; Vectors;
Conference_Titel :
Information Theory, 1994. Proceedings., 1994 IEEE International Symposium on
Conference_Location :
Trondheim
Print_ISBN :
0-7803-2015-8
DOI :
10.1109/ISIT.1994.394752
