Han-gyoo Kim

EECS Department
University of California, Berkeley
Technical Report No. UCB/CSD-92-677
January 1992

http://www2.eecs.berkeley.edu/Pubs/TechRpts/1992/CSD-92-677.pdf

This paper at first applies Stone-Weierstrass theorem to prove the existence of universal approximating capability of feedforward multilayer neural nets, and at second discusses the construction of such approximators. Since the fact that approximators can be built efficiently without using neural net structures provides insight of great importance into understanding the characteristics of neural nets, this paper further investigates the limit of neural net capability. Finally, feasibility of utilizing general approximators to solve AI problems is discussed.

BibTeX citation:

@techreport{Kim:CSD-92-677,
    Author = {Kim, Han-gyoo},
    Title = {Mathematical Findings on Artificial Neural Nets and Their Physical Meaning},
    Institution = {EECS Department, University of California, Berkeley},
    Year = {1992},
    Month = {Jan},
    URL = {http://www2.eecs.berkeley.edu/Pubs/TechRpts/1992/5436.html},
    Number = {UCB/CSD-92-677},
    Abstract = {This paper at first applies Stone-Weierstrass theorem to prove the existence of universal approximating capability of feedforward multilayer neural nets, and at second discusses the construction of such approximators. Since the fact that approximators can be built efficiently without using neural net structures provides insight of great importance into understanding the characteristics of neural nets, this paper further investigates the limit of neural net capability. Finally, feasibility of utilizing general approximators to solve AI problems is discussed.}
}

EndNote citation:

%0 Report
%A Kim, Han-gyoo
%T Mathematical Findings on Artificial Neural Nets and Their Physical Meaning
%I EECS Department, University of California, Berkeley
%D 1992
%@ UCB/CSD-92-677
%U http://www2.eecs.berkeley.edu/Pubs/TechRpts/1992/5436.html
%F Kim:CSD-92-677