Adaptive quadratic Neural Nets.

S.G. Lim; Michael Alder; P.T. Hadingham

doi:10.1016/0167-8655(92)90029-Y

Adaptive quadratic Neural Nets.

S.G. Lim, Michael Alder, P.T. Hadingham

School of Molecular Sciences

Research output: Contribution to journal › Article › peer-review

12 Citations (Web of Science)

Abstract

We present the theory and some results of a new algorithm for Artificial Neural Nets which behaves well on complex data sets. The algorithm uses adaptive quadratic forms as discriminant functions and is very fast compared with Back-Propagation - improvements of four orders of magnitude have been obtained.

Original language	English
Pages (from-to)	325-329
Journal	Pattern Recognition Letters
Volume	13
Issue number	May, 1992
DOIs	https://doi.org/10.1016/0167-8655(92)90029-Y
Publication status	Published - 1992

Access to Document

10.1016/0167-8655(92)90029-Y

Cite this

@article{8c428224b0f945b5b438f17c63100d3e,

title = "Adaptive quadratic Neural Nets.",

abstract = "We present the theory and some results of a new algorithm for Artificial Neural Nets which behaves well on complex data sets. The algorithm uses adaptive quadratic forms as discriminant functions and is very fast compared with Back-Propagation - improvements of four orders of magnitude have been obtained.",

author = "S.G. Lim and Michael Alder and P.T. Hadingham",

year = "1992",

doi = "10.1016/0167-8655(92)90029-Y",

language = "English",

volume = "13",

pages = "325--329",

journal = "Pattern Recogniton Letters",

issn = "0167-8655",

publisher = "Elsevier",

number = "May, 1992",

}

TY - JOUR

T1 - Adaptive quadratic Neural Nets.

AU - Lim, S.G.

AU - Alder, Michael

AU - Hadingham, P.T.

PY - 1992

Y1 - 1992

N2 - We present the theory and some results of a new algorithm for Artificial Neural Nets which behaves well on complex data sets. The algorithm uses adaptive quadratic forms as discriminant functions and is very fast compared with Back-Propagation - improvements of four orders of magnitude have been obtained.

AB - We present the theory and some results of a new algorithm for Artificial Neural Nets which behaves well on complex data sets. The algorithm uses adaptive quadratic forms as discriminant functions and is very fast compared with Back-Propagation - improvements of four orders of magnitude have been obtained.

U2 - 10.1016/0167-8655(92)90029-Y

DO - 10.1016/0167-8655(92)90029-Y

M3 - Article

VL - 13

SP - 325

EP - 329

JO - Pattern Recogniton Letters

JF - Pattern Recogniton Letters

SN - 0167-8655

IS - May, 1992

ER -

Adaptive quadratic Neural Nets.

Abstract

Access to Document

Fingerprint

Cite this