Critical points for neural net least-squares problems

Research output: Contribution to conferencePaperpeer-review

2 Scopus citations

Abstract

For least-squares problems involving parameterized analytic functions, this paper establishes generic countability, and under stronger assumptions finiteness, of critical points of the quadratic loss function. For single-hidden layer sigmoidal neural networks, an upper bound is provided.

Original languageEnglish (US)
Pages2949-2954
Number of pages6
StatePublished - 1995
EventProceedings of the 1995 IEEE International Conference on Neural Networks. Part 1 (of 6) - Perth, Aust
Duration: Nov 27 1995Dec 1 1995

Other

OtherProceedings of the 1995 IEEE International Conference on Neural Networks. Part 1 (of 6)
CityPerth, Aust
Period11/27/9512/1/95

All Science Journal Classification (ASJC) codes

  • Software

Fingerprint

Dive into the research topics of 'Critical points for neural net least-squares problems'. Together they form a unique fingerprint.

Cite this