With this factor, a backpropagation algorithm based on single parameter is sound and quick.
针对前馈神经网络应时变输入的自学习机制 ,采用李雅普诺夫函数来分析权值的收敛性 ,从而揭示 BP神经网络算法朝最小误差方向调整权值的内在因素 ,并在分析单参数 BP算法收敛性基础上 ,提出单参数变调整法则的离散型 BP神经网络算法 。
Single parameter quick search method, an improved BP algorithm based on single parameter coordinate rotated method, is presented in this paper.
提出了一种基于单参数坐标轮换法的改进 BP算法 。
Based on a fibre-bundle surface model using one-parameter groups of a linear Lie algebra as fibres is introduced, a new extended model is proposed using a fibre-bundle of one-parameter groups of a Hamiltonian Lie algebra, which can be used for identification of 3D objects.
介绍一种以线性李代数的单参数群为纤维的纤维丛曲面;引进一种新的以哈密顿李代数为纤维的曲面模型,该模型可以用于3D物体识别;得出此曲面模型的高斯曲率,并给出实例。