دانلود رایگان مقاله انگلیسی تحلیل بی ثباتی و پایداری چندگانه شبکه های عصبی بازگردنده با تاخیر های زمانی مختلف - الزویر 2018

عنوان فارسی
تجزیه و تحلیل بی ثباتی و پایداری چندگانه شبکه های عصبی بازگردنده با تاخیر های زمانی مختلف
عنوان انگلیسی
Multistability and instability analysis of recurrent neural networks with time-varying delays
صفحات مقاله فارسی
0
صفحات مقاله انگلیسی
11
سال انتشار
2018
نشریه
الزویر - Elsevier
فرمت مقاله انگلیسی
PDF
کد محصول
E6429
رشته های مرتبط با این مقاله
کامپیوتر، فناوری اطلاعات
گرایش های مرتبط با این مقاله
شبکه های کامپیوتری
مجله
شبکه های عصبی - Neural Networks
دانشگاه
School of Automation - Huazhong University of Science and Technology - China
کلمات کلیدی
شبکه عصبی مکرر، ثبات چندگانه، بی ثباتی، تاخیر زمانی متغیر
چکیده

abstract


This paper provides new theoretical results on the multistability and instability analysis of recurrent neural networks with time-varying delays. It is shown that such n-neuronal recurrent neural networks have exactly (4k + 3)k0 equilibria, (2k + 2)k0 of which are locally exponentially stable and the others are unstable, where k0 is a nonnegative integer such that k0 ≤ n. By using the combination method of two different divisions, recurrent neural networks can possess more dynamic properties. This method improves and extends the existing results in the literature. Finally, one numerical example is provided to show the superiority and effectiveness of the presented results.

نتیجه گیری

5. Conclusion


In this paper, we have discussed the multistability and instability issue of delayed recurrent neural networks. By the division of state space and the dimensional space reconstruction, some sufficient criteria have been established to ensure the existence of (2k + 2)k0 locally exponentially stable equilibria, and (4k + 3)k0 − (2k+2)k0 equilibria are unstable, where k0 is a nonnegative integer such that k0 ≤ n. These new criteria improve and extend the existing results of multistability in the literature. By means of coupling division, it also reveals that the different regions of parameter are influenced by division, and these regions of parameter are allowed to have more options, in which the dynamic behaviors are more abundant. Finally, a numerical simulation is conducted to illustrate the derived theoretical results.


بدون دیدگاه