Design of High-Capacity Auto-Associative Memories Based on the Analysis of Complex-Valued Neural Networks

被引：0

作者：

Gong, Weiqiang ^{[1
]}

Liang, Jinling ^{[2
]}

Li, Dehao ^{[2
]}

机构：

[1] Nanjing Univ Finance & Econ, Sch Appl Math, Nanjing 210023, Jiangsu, Peoples R China

[2] Southeast Univ, Sch Math, Nanjing 210096, Jiangsu, Peoples R China

来源：

2017 14TH INTERNATIONAL WORKSHOP ON COMPLEX SYSTEMS AND NETWORKS (IWCSN) | 2017年

关键词：

complex-valued neural networks; auto-associative memory; mixed delays; exponential stability; store pattern; ASSOCIATIVE MEMORY; STABILITY; BIFURCATION; DELAYS; MODEL;

D O I：

暂无

中图分类号：

TP301 [理论、方法];

学科分类号：

081202 ;

摘要：

This paper presents a new design method for synthesizing associative memories based on the complex-valued neural networks with asynchronous and unbounded delays. Some sufficient criteria in terms of several inequalities, which can bring robustness to parameters of the complex-valued networks, are obtained to assure the existence, uniqueness and global exponential stability of equilibrium points for the addressed networks with mixed delays. The design procedure enables auto-associative memories to be synthesized with high storage capacity. The desired store patterns are retrieved by feeding proper probes through external inputs rather than initial conditions, which can avoid spurious memory patterns. The obtained results are much more general than the existed works. Finally, one numerical example is provided to illustrate the effectiveness of the obtained results.

引用

页码：161 / 168

页数：8

共 26 条

[1] Less conservative delay-dependent H∞ control of uncertain neural networks with discrete interval and distributed time-varying delays [J].

Ali, M. Syed ;

Saravanakumar, R. ;

Zhu, Quanxin .

NEUROCOMPUTING, 2015, 166 :84-95

[2] Single-layered complex-valued neural network for real-valued classification problems [J].

Amin, Md. Faijul ;

Murase, Kazuyuki .

NEUROCOMPUTING, 2009, 72 (4-6) :945-955

[3]

Aoki H, 2000, ELECTRON COMM JPN 3, V83, P62, DOI 10.1002/(SICI)1520-6440(200007)83:7<62::AID-ECJC7>3.0.CO

[4]

2-L

[5] CHANNEL EQUALIZATION USING ADAPTIVE COMPLEX RADIAL BASIS FUNCTION NETWORKS [J].

CHA, I ;

KASSAM, SA .

IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS, 1995, 13 (01) :122-131

[6] Stability and Hopf bifurcation of a complex-valued neural network with two time delays [J].

Dong, Tao ;

Liao, Xiaofeng ;

Wang, Aijuan .

NONLINEAR DYNAMICS, 2015, 82 (1-2) :173-184

[7] Chaotic complex-valued bidirectional associative memory with a real-valued context part [J].

Furusawa, Shogo ;

Kobayashi, Masaki .

IEICE NONLINEAR THEORY AND ITS APPLICATIONS, 2013, 4 (03) :299-312

[8] An augmented extended Kalman filter algorithm for complex-valued recurrent neural networks [J].

Goh, Su Lee ;

Mandic, Danilo P. .

NEURAL COMPUTATION, 2007, 19 (04) :1039-1055

[9] On discrete-time cellular neural networks for associative memories [J].

Grassi, G .

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-FUNDAMENTAL THEORY AND APPLICATIONS, 2001, 48 (01) :107-111

[10] Analysis and design of associative memories based on stability of cellular neural networks [J].

Han, Qi ;

Liao, Xiaofeng ;

Huang, Tingwen ;

Peng, Jun ;

Li, Chuandong ;

Huang, Hongyu .

NEUROCOMPUTING, 2012, 97 :192-200

← 1 2 3 →