Accession Number:

ADP013757

Title:

Approximation by Perceptron Networks

Descriptive Note:

Conference paper

Corporate Author:

CZECH ACADEMY OF SCIENCES PRAGUE INST OF COMPUTER SCIENCE

Personal Author(s):

Report Date:

2001-07-01

Pagination or Media Count:

7.0

Abstract:

In this paper we survey our recent results on properties of approximation by linear combinations of characteristic functions of half-spaces. We focus on existence of best approximation, impossibility of choosing among best approximations a continuous one, estimates of rates of approximation by linear combinations of n characteristic functions of half-spaces and integral representation as a linear combination of a continuum of half-spaces.

Subject Categories:

  • Numerical Mathematics
  • Theoretical Mathematics

Distribution Statement:

APPROVED FOR PUBLIC RELEASE