Comparison of complex- and real-valued feedforward neural networks in their generalization ability

Akira Hirose; Shotaro Yoshida

doi:10.1007/978-3-642-24955-6_63

Comparison of complex- and real-valued feedforward neural networks in their generalization ability

Akira Hirose, Shotaro Yoshida

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

10 Citations (Scopus)

Abstract

We compare the generalization characteristics of complex-valued and real-valued feedforward neural networks when they deal with wave-related signals. We assume a task of function approximation. Experiments demonstrate that complex-valued neural networks show smaller generalization error than real-valued ones in particular when the signals have high degree of wave nature.

Original language	English
Title of host publication	Neural Information Processing - 18th International Conference, ICONIP 2011, Proceedings
Pages	526-531
Number of pages	6
Edition	PART 1
DOIs	https://doi.org/10.1007/978-3-642-24955-6_63
Publication status	Published - 2011 Nov 28
Externally published	Yes
Event	18th International Conference on Neural Information Processing, ICONIP 2011 - Shanghai, China Duration: 2011 Nov 13 → 2011 Nov 17

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Number	PART 1
Volume	7062 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Other

Other	18th International Conference on Neural Information Processing, ICONIP 2011
Country/Territory	China
City	Shanghai
Period	11/11/13 → 11/11/17

Keywords

Complex-valued neural network
function approximation
generalization

ASJC Scopus subject areas

Computer Science(all)
Theoretical Computer Science

Access to Document

10.1007/978-3-642-24955-6_63

Cite this

Hirose, A., & Yoshida, S. (2011). Comparison of complex- and real-valued feedforward neural networks in their generalization ability. In Neural Information Processing - 18th International Conference, ICONIP 2011, Proceedings (PART 1 ed., pp. 526-531). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7062 LNCS, No. PART 1). https://doi.org/10.1007/978-3-642-24955-6_63

Comparison of complex- and real-valued feedforward neural networks in their generalization ability. / Hirose, Akira; Yoshida, Shotaro.

Neural Information Processing - 18th International Conference, ICONIP 2011, Proceedings. PART 1. ed. 2011. p. 526-531 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7062 LNCS, No. PART 1).

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

Hirose, A & Yoshida, S 2011, Comparison of complex- and real-valued feedforward neural networks in their generalization ability. in Neural Information Processing - 18th International Conference, ICONIP 2011, Proceedings. PART 1 edn, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), no. PART 1, vol. 7062 LNCS, pp. 526-531, 18th International Conference on Neural Information Processing, ICONIP 2011, Shanghai, China, 11/11/13. https://doi.org/10.1007/978-3-642-24955-6_63

Hirose A, Yoshida S. Comparison of complex- and real-valued feedforward neural networks in their generalization ability. In Neural Information Processing - 18th International Conference, ICONIP 2011, Proceedings. PART 1 ed. 2011. p. 526-531. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 1). https://doi.org/10.1007/978-3-642-24955-6_63

Hirose, Akira ; Yoshida, Shotaro. / Comparison of complex- and real-valued feedforward neural networks in their generalization ability. Neural Information Processing - 18th International Conference, ICONIP 2011, Proceedings. PART 1. ed. 2011. pp. 526-531 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 1).

@inproceedings{f519e8176f0a45f3a95335bbb6fa7ad1,

title = "Comparison of complex- and real-valued feedforward neural networks in their generalization ability",

abstract = "We compare the generalization characteristics of complex-valued and real-valued feedforward neural networks when they deal with wave-related signals. We assume a task of function approximation. Experiments demonstrate that complex-valued neural networks show smaller generalization error than real-valued ones in particular when the signals have high degree of wave nature.",

keywords = "Complex-valued neural network, function approximation, generalization",

author = "Akira Hirose and Shotaro Yoshida",

year = "2011",

month = nov,

day = "28",

doi = "10.1007/978-3-642-24955-6_63",

language = "English",

isbn = "9783642249549",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

number = "PART 1",

pages = "526--531",

booktitle = "Neural Information Processing - 18th International Conference, ICONIP 2011, Proceedings",

edition = "PART 1",

note = "18th International Conference on Neural Information Processing, ICONIP 2011 ; Conference date: 13-11-2011 Through 17-11-2011",

}

TY - GEN

T1 - Comparison of complex- and real-valued feedforward neural networks in their generalization ability

AU - Hirose, Akira

AU - Yoshida, Shotaro

PY - 2011/11/28

Y1 - 2011/11/28

N2 - We compare the generalization characteristics of complex-valued and real-valued feedforward neural networks when they deal with wave-related signals. We assume a task of function approximation. Experiments demonstrate that complex-valued neural networks show smaller generalization error than real-valued ones in particular when the signals have high degree of wave nature.

AB - We compare the generalization characteristics of complex-valued and real-valued feedforward neural networks when they deal with wave-related signals. We assume a task of function approximation. Experiments demonstrate that complex-valued neural networks show smaller generalization error than real-valued ones in particular when the signals have high degree of wave nature.

KW - Complex-valued neural network

KW - function approximation

KW - generalization

UR - http://www.scopus.com/inward/record.url?scp=81855190753&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=81855190753&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-24955-6_63

DO - 10.1007/978-3-642-24955-6_63

M3 - Conference contribution

AN - SCOPUS:81855190753

SN - 9783642249549

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 526

EP - 531

BT - Neural Information Processing - 18th International Conference, ICONIP 2011, Proceedings

T2 - 18th International Conference on Neural Information Processing, ICONIP 2011

Y2 - 13 November 2011 through 17 November 2011

ER -

Comparison of complex- and real-valued feedforward neural networks in their generalization ability

Abstract

Publication series

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this