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

Authors:
Akira Hirose;Shotaro Yoshida
Affiliations:
Department of Electrical Engineering and Information Systems, The University of Tokyo, Japan;Department of Electrical Engineering and Information Systems, The University of Tokyo, Japan
Venue:
ICONIP'11 Proceedings of the 18th international conference on Neural Information Processing - Volume Part I
Year:
2011

Citing 4
Cited 0

Applications of complex-valued neural networks to coherent optical computing using phase-sensitive detection scheme

Information Sciences—Applications: An International Journal
An extension of the back-propagation algorithm to complex numbers

Neural Networks
Complex-Valued Neural Networks (Studies in Computational Intelligence)

Complex-Valued Neural Networks (Studies in Computational Intelligence)
Behavior control of coherent-type neural networks by carrier-frequency modulation

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.00

Visualization

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.