Polynomial Bounds for VC Dimension of Sigmoidal Neural Networks

TitlePolynomial Bounds for VC Dimension of Sigmoidal Neural Networks
Publication TypeTechnical Report
Year of Publication1995
AuthorsKarpinski, M., & Macintyre A.
Other Numbers941
Abstract

We introduce a new method for proving explicit upper bounds on the VC Dimension of general functional basis networks, and prove as an application, for the first time, the VC Dimension of analog neural networks with the sigmoid activation function ?(y)=1/1+e^{-y} to be bounded by a quadratic polynomial in the number of programmable parameters.

URLhttp://www.icsi.berkeley.edu/ftp/global/pub/techreports/1995/tr-95-001.pdf
Bibliographic Notes

ICSI Technical Report TR-95-001

Abbreviated Authors

M. Karpinski and A. Macintyre

ICSI Publication Type

Technical Report