On Chaotic Neural Network Design: A New Framework

详细信息    查看全文

• 作者：Ke Qin
• 关键词：Chaotic neural networks ; Chaos ; Nonlinear dynamics ; Self ; feedback ; Recurrent neural networks
• 刊名：Neural Processing Letters
• 出版年：2017
• 出版时间：February 2017
• 年：2017
• 卷：45
• 期：1
• 页码：243-261
• 全文大小：
• 刊物类别：Physics and Astronomy
• 刊物主题：Artificial Intelligence (incl. Robotics); Complex Systems; Computational Intelligence;
• 出版者：Springer US
• ISSN：1573-773X
• 卷排序：45

文摘

The theory of chaos and chaotic neural networks (CNNs) has been widely investigated in the past two decades. However, most researchers in this area have merely focused on how to make full use of CNNs to solve various problems in areas such as pattern recognition, classification, associate memory and cryptography. The philosophy of how to design a CNN is seldom discussed. In this paper, we present a general methodology for designing CNNs. By appropriately choosing a self-feedback mechanism, and also including coupling functions and an external stimulus, we have succeeded in proving that a dynamical system, defined by discrete time feedback equations, does, indeed, possess interesting chaotic properties. To the best of our knowledge, the results presented here are novel and pioneering.