作者: Ziye Zhang , Xiaoping Liu , Jian Chen , Runan Guo , Shaowei Zhou
DOI: 10.1016/J.NEUCOM.2017.04.013
关键词:
摘要: Abstract This paper focuses on the stability problem for delayed complex-valued recurrent neural networks. Whether activation functions are explicitly expressed by separating real and imaginary parts or not, they always assumed to satisfy globally Lipschitz condition in complex domain. For two cases of functions, based homeomorphism theory Lyapunov function approach new delay-dependent sufficient conditions guarantee existence, uniqueness, asymptotical equilibrium point system obtained, respectively. each case, several numerical examples given show effectiveness advantages obtained results.