Complex valued neural networks for the solution of the two-spiral problem

Wael R. Elwasif; Laurene V. Fausett

Complex valued neural networks for the solution of the two-spiral problem

Research output: Contribution to conference › Paper › peer-review

Abstract

For the difficult problem of separating two concentric spirals, a significant improvement in learning and generalization is demonstrated for backpropagation networks using complex, rather than real, signals and weights.

Original language	English
Pages	227-232
Number of pages	6
State	Published - 1995
Externally published	Yes
Event	Proceedings of the 1995 Artificial Neural Networks in Engineering, ANNIE'95 - St.Louis, MO, USA Duration: Nov 12 1995 → Nov 15 1995

Conference

Conference	Proceedings of the 1995 Artificial Neural Networks in Engineering, ANNIE'95
City	St.Louis, MO, USA
Period	11/12/95 → 11/15/95

Cite this

@conference{b7bdd6873fe44101a4c730dd0fbd4819,

title = "Complex valued neural networks for the solution of the two-spiral problem",

abstract = "For the difficult problem of separating two concentric spirals, a significant improvement in learning and generalization is demonstrated for backpropagation networks using complex, rather than real, signals and weights.",

author = "Elwasif, {Wael R.} and Fausett, {Laurene V.}",

year = "1995",

language = "English",

pages = "227--232",

note = "Proceedings of the 1995 Artificial Neural Networks in Engineering, ANNIE'95 ; Conference date: 12-11-1995 Through 15-11-1995",

}

TY - CONF

T1 - Complex valued neural networks for the solution of the two-spiral problem

AU - Elwasif, Wael R.

AU - Fausett, Laurene V.

PY - 1995

Y1 - 1995

N2 - For the difficult problem of separating two concentric spirals, a significant improvement in learning and generalization is demonstrated for backpropagation networks using complex, rather than real, signals and weights.

AB - For the difficult problem of separating two concentric spirals, a significant improvement in learning and generalization is demonstrated for backpropagation networks using complex, rather than real, signals and weights.

UR - http://www.scopus.com/inward/record.url?scp=0029417743&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:0029417743

SP - 227

EP - 232

T2 - Proceedings of the 1995 Artificial Neural Networks in Engineering, ANNIE'95

Y2 - 12 November 1995 through 15 November 1995

ER -

Complex valued neural networks for the solution of the two-spiral problem

Abstract

Conference

Other files and links

Fingerprint

Cite this