Nonlinearly Activated Recurrent Neural Network for Computing the Drazin Inverse
作者: Xue-Zhong WangHaifeng MaPredrag S. Stanimirović
作者单位: 1Hexi University
2Harbin Normal University
3University of Niš
刊名: Neural Processing Letters, 2017, Vol.46 (1), pp.195-217
来源数据库: Springer Nature Journal
DOI: 10.1007/s11063-017-9581-y
关键词: Recurrent neural networkDrazin inverseDynamic equationActivation function15A0965F20
英文摘要: Four gradient-based recurrent neural networks for computing the Drazin inverse of a square real matrix are developed. Theoretical analysis shows that any monotonically-increasing odd activation function ensures the global convergence performance of defined neural network models. The computer simulation results further substantiate that the considered neural networks could compute the Drazin inverse with accuracy and effectiveness. Moreover, the presented neural networks show superior convergence in the case when the power-sigmoid activation functions are used compared to linear models.
原始语种摘要: Four gradient-based recurrent neural networks for computing the Drazin inverse of a square real matrix are developed. Theoretical analysis shows that any monotonically-increasing odd activation function ensures the global convergence performance of defined neural network models. The computer simulation results further substantiate that the considered neural networks could compute the Drazin inverse with accuracy and effectiveness. Moreover, the presented neural networks show superior convergence in the case when the power-sigmoid activation functions are used compared to linear models.
全文获取路径: Springer Nature  (合作)
分享到:
来源刊物:
影响因子:1.24 (2012)

×
关键词翻译
关键词翻译
  • neural 神经系统的
  • sigmoid s状的
  • network 网络
  • computer 电子计算机
  • inverse 逆的
  • effectiveness 有效性
  • convergence 汇合
  • square 正方形
  • gradient 倾斜度
  • substantiate 证明