arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1503.01818math.OC
收藏

离散时间循环神经网络和非线性最优化问题的稳定性

Stability of Discrete time Recurrent Neural Networks and Nonlinear optimization problems

Nikita Barabanov, Jayant Singh

摘要:

本文利用耗散域约化方法证明了离散时间循环神经网络的全局李雅普诺夫稳定性。这些本质上为非线性系统的A稳定性的标准和先进的准则产生相当弱的结果。实践证明,该方法具有较强的实用性。它涉及到一个多步骤的过程,在每一步上的特殊非凸函数的最大化。我们得到的条件,保证存在最多一个点的局部极大值,这样的功能在每个超平面。这个非平凡的结果对于广泛的神经元传递函数是有效的。

英文摘要:

We consider the method of Reduction of Dissipativity Domain to prove global Lyapunov stability of Discrete Time Recurrent Neural Networks. The standard and advanced criteria for Absolute Stability of these essentially nonlinear systems produce rather weak results. The method mentioned above is proved to be more powerful. It involves a multi-step procedure with maximization of special nonconvex functions over polytopes on every step. We derive conditions which guarantee an existence of at most one point of local maximum for such functions over every hyperplane. This nontrivial result is valid for wide range of neuron transfer functions.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑