On the infeasibility of training neural networks with small mean-squared error

Authors
Citation
Vh. Vu, On the infeasibility of training neural networks with small mean-squared error, IEEE INFO T, 44(7), 1998, pp. 2892-2900
Citations number
12
Categorie Soggetti
Information Tecnology & Communication Systems
Journal title
IEEE TRANSACTIONS ON INFORMATION THEORY
ISSN journal
00189448 → ACNP
Volume
44
Issue
7
Year of publication
1998
Pages
2892 - 2900
Database
ISI
SICI code
0018-9448(199811)44:7<2892:OTIOTN>2.0.ZU;2-H
Abstract
We demonstrate that the problem of training neural networks with small mean -squared error is computationally intractable. This answers a question pose d by L. Jones [6].