COMPLEX-DOMAIN BACKPROPAGATION

被引：246

作者：

GEORGIOU, GM

KOUTSOUGERAS, C

机构：

[1] Computer Science Department, Tulane University, New Orleans, LA

来源：

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-ANALOG AND DIGITAL SIGNAL PROCESSING | 1992年 / 39卷 / 05期

关键词：

D O I：

10.1109/82.142037

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

The well-known backpropagation algorithm is extended to complex domain backpropagation (CDBP) which can be used to train neural networks for which the inputs, weights, activation functions, and outputs are complex-valued. Previous derivations of CDBP were necessarily admitting activation functions that have singularities, which is highly undesirable. Here CDBP is derived so that it accommodates classes of suitable activation functions. One such function is found and the circuit implementation of the corresponding neuron is given. CDBP hardware circuits can be used to process sinusoidal signals all at the same frequency (phasors).

引用

页码：330 / 334

页数：5

共 12 条

[1]

[Anonymous], 1987, LEARNING INTERNAL RE

[2]

[Anonymous], 1992, NEURAL NETWORKS FUZZ

[3]

Churchill R. V., 1948, COMPLEX VARIABLES AP, V1st

[4] ADAPTIVE FILTERING IN FREQUENCY-DOMAIN [J].

DENTINO, M ;

MCCOOL, J ;

WIDROW, B .

PROCEEDINGS OF THE IEEE, 1978, 66 (12) :1658-1659

[5]

KIM MS, 1990, IJCNN INT JOINT C NE

[6] THE COMPLEX BACKPROPAGATION ALGORITHM [J].

LEUNG, H ;

HAYKIN, S .

IEEE TRANSACTIONS ON SIGNAL PROCESSING, 1991, 39 (09) :2101-2104

[7]

Millman J, 1979, MICROELECTRONICS

[8]

STOCKWELL EB, 1991, TECH REP ASTRONOMY U

[9]

THRIFT P, 1990, TEXAS INSTRUMENTS TE, V7

[10] COMPLEX LMS ALGORITHM [J].

WIDROW, B ;

MCCOOL, J ;

BALL, M .

PROCEEDINGS OF THE IEEE, 1975, 63 (04) :719-720

← 1 2 →