View PDF
Access through your institution
Neural Networks
Volume 2, Issue 5, 1989, Pages 359-366
Original contribution
Multilayer feedforward networks are universal approximators
Get rights and content
This paper rigorously establishes that standard multilayer feedforward networks with as few as one hidden layer using arbitrary squashing functions are capable of approximating any Borel measurable function from one finite dimensional space to another to any desired degree of accuracy, provided sufficiently many hidden units are available. In this sense, multilayer feedforward networks are a class of universal approximators.
Feedforward networksUniversal approximation​Mapping networksNetwork representation capability​Stone-Weierstrass Theorem​Squashing functionsSigma-Pi networks​Back-propagation networks
White's participation was supported by a grant from the Guggenheim Foundation and by National Science Foundation Grant SES-8806990. The authors are grateful for helpful suggestions by the referees.
View full text
Copyright © 1989 Published by Elsevier Ltd.
About ScienceDirect
Remote access
Shopping cart
Contact and support
Terms and conditions
Privacy policy
We use cookies to help provide and enhance our service and tailor content and ads. By continuing you agree to the use of cookies.
Copyright © 2021 Elsevier B.V. or its licensors or contributors. ScienceDirect ® is a registered trademark of Elsevier B.V.
ScienceDirect ® is a registered trademark of Elsevier B.V.
Journals & Books
Journals & Books