摘要
给出了延迟离散Ｈｏｐｆｉｅｌｄ－型神经网络的动态特征，提出了新的异步运行规则，证明了非对称连接权阵条件下的收敛性定理，推广了已有的延迟离散Ｈｏｐｆｉｅｌｄ－型神经网络的收敛性结果．
To provide dynamic characteristics on discrete Asymmetric Hopfield\|type neural networks with delay, a new operating mode is presented. We prove that any discrete Hopfield\|type neural networks with delay will converge to stable state when operating in the serial mode under a proper assumption, whose weight matrix is a asymmetric and generalize convergence theorem in earlier works. We also relate maximum of modified energy function with stable state of neural networks with delay.
出处
《大庆石油学院学报》
CAS
北大核心
1999年第2期39-43,共5页
Journal of Daqing Petroleum Institute
基金
国家自然科学基金
关键词
延迟神经网络
收敛性
稳定态
Hopfield-型
异步运行
neural networks with delay, convergence, stable state,Hopfield\|type, operating in the serial mode
