A new class of high-order neural networks with nonlinear decision boundaries

A. Bouzerdoum

doi:10.1109/ICONIP.1999.844673

A new class of high-order neural networks with nonlinear decision boundaries

A. Bouzerdoum^*

^*Corresponding author for this work

Edith Cowan University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

21 Citations (Scopus)

Abstract

Presents a class of high-order neural networks called shunting inhibitory artificial neural networks (SIANNs) for classification and function approximation tasks. In these networks, the basic synaptic interaction is of the shunting inhibitory type. Due to the nonlinearity mediated by shunting inhibition, these networks are capable of producing classifiers with complex nonlinear decision boundaries, ranging from simple hyperplanes to very complex nonlinear surfaces. Therefore, developing efficient training algorithms for these networks will simplify the design of very powerful classifiers and function approximators. In this paper, we present a training method for a feedforward SIANN based on the backpropagation algorithm and on gradient descent.

Original language	English
Title of host publication	ICONIP 1999, 6th International Conference on Neural Information Processing - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1004-1009
Number of pages	6
ISBN (Electronic)	0780358716, 9780780358713
DOIs	https://doi.org/10.1109/ICONIP.1999.844673
Publication status	Published - 1999
Externally published	Yes
Event	6th International Conference on Neural Information Processing, ICONIP 1999 - Perth, Australia Duration: 16 Nov 1999 → 20 Nov 1999

Publication series

Name	ICONIP 1999, 6th International Conference on Neural Information Processing - Proceedings
Volume	3

Conference

Conference	6th International Conference on Neural Information Processing, ICONIP 1999
Country/Territory	Australia
City	Perth
Period	16/11/99 → 20/11/99

Access to Document

10.1109/ICONIP.1999.844673

Cite this

Bouzerdoum, A. (1999). A new class of high-order neural networks with nonlinear decision boundaries. In ICONIP 1999, 6th International Conference on Neural Information Processing - Proceedings (pp. 1004-1009). Article 844673 (ICONIP 1999, 6th International Conference on Neural Information Processing - Proceedings; Vol. 3). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICONIP.1999.844673

@inproceedings{ca0dffaf23ab446bb9fdc4ab67eaa199,

title = "A new class of high-order neural networks with nonlinear decision boundaries",

abstract = "Presents a class of high-order neural networks called shunting inhibitory artificial neural networks (SIANNs) for classification and function approximation tasks. In these networks, the basic synaptic interaction is of the shunting inhibitory type. Due to the nonlinearity mediated by shunting inhibition, these networks are capable of producing classifiers with complex nonlinear decision boundaries, ranging from simple hyperplanes to very complex nonlinear surfaces. Therefore, developing efficient training algorithms for these networks will simplify the design of very powerful classifiers and function approximators. In this paper, we present a training method for a feedforward SIANN based on the backpropagation algorithm and on gradient descent.",

author = "A. Bouzerdoum",

note = "Publisher Copyright: {\textcopyright} 1999 IEEE.; 6th International Conference on Neural Information Processing, ICONIP 1999 ; Conference date: 16-11-1999 Through 20-11-1999",

year = "1999",

doi = "10.1109/ICONIP.1999.844673",

language = "English",

series = "ICONIP 1999, 6th International Conference on Neural Information Processing - Proceedings",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "1004--1009",

booktitle = "ICONIP 1999, 6th International Conference on Neural Information Processing - Proceedings",

address = "United States",

}

Bouzerdoum, A 1999, A new class of high-order neural networks with nonlinear decision boundaries. in ICONIP 1999, 6th International Conference on Neural Information Processing - Proceedings., 844673, ICONIP 1999, 6th International Conference on Neural Information Processing - Proceedings, vol. 3, Institute of Electrical and Electronics Engineers Inc., pp. 1004-1009, 6th International Conference on Neural Information Processing, ICONIP 1999, Perth, Australia, 16/11/99. https://doi.org/10.1109/ICONIP.1999.844673

A new class of high-order neural networks with nonlinear decision boundaries. / Bouzerdoum, A.
ICONIP 1999, 6th International Conference on Neural Information Processing - Proceedings. Institute of Electrical and Electronics Engineers Inc., 1999. p. 1004-1009 844673 (ICONIP 1999, 6th International Conference on Neural Information Processing - Proceedings; Vol. 3).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - A new class of high-order neural networks with nonlinear decision boundaries

AU - Bouzerdoum, A.

PY - 1999

Y1 - 1999

N2 - Presents a class of high-order neural networks called shunting inhibitory artificial neural networks (SIANNs) for classification and function approximation tasks. In these networks, the basic synaptic interaction is of the shunting inhibitory type. Due to the nonlinearity mediated by shunting inhibition, these networks are capable of producing classifiers with complex nonlinear decision boundaries, ranging from simple hyperplanes to very complex nonlinear surfaces. Therefore, developing efficient training algorithms for these networks will simplify the design of very powerful classifiers and function approximators. In this paper, we present a training method for a feedforward SIANN based on the backpropagation algorithm and on gradient descent.

AB - Presents a class of high-order neural networks called shunting inhibitory artificial neural networks (SIANNs) for classification and function approximation tasks. In these networks, the basic synaptic interaction is of the shunting inhibitory type. Due to the nonlinearity mediated by shunting inhibition, these networks are capable of producing classifiers with complex nonlinear decision boundaries, ranging from simple hyperplanes to very complex nonlinear surfaces. Therefore, developing efficient training algorithms for these networks will simplify the design of very powerful classifiers and function approximators. In this paper, we present a training method for a feedforward SIANN based on the backpropagation algorithm and on gradient descent.

UR - https://www.scopus.com/pages/publications/0038294101

U2 - 10.1109/ICONIP.1999.844673

DO - 10.1109/ICONIP.1999.844673

M3 - Conference contribution

AN - SCOPUS:0038294101

T3 - ICONIP 1999, 6th International Conference on Neural Information Processing - Proceedings

SP - 1004

EP - 1009

BT - ICONIP 1999, 6th International Conference on Neural Information Processing - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 6th International Conference on Neural Information Processing, ICONIP 1999

Y2 - 16 November 1999 through 20 November 1999

ER -

Bouzerdoum A. A new class of high-order neural networks with nonlinear decision boundaries. In ICONIP 1999, 6th International Conference on Neural Information Processing - Proceedings. Institute of Electrical and Electronics Engineers Inc. 1999. p. 1004-1009. 844673. (ICONIP 1999, 6th International Conference on Neural Information Processing - Proceedings). doi: 10.1109/ICONIP.1999.844673

A new class of high-order neural networks with nonlinear decision boundaries

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this