摘要
详细研究了求解病态线性方程组AX=b的神经网络算法,提出并证明了神经网络算法的收敛性定理,算法的收敛性定理为神经网络学习率的选择提供了理论依据.为了验证算法的有效性,给出了应用实例.研究结果表明了求解病态线性方程组的神经网络算法是有效的.
The neural-network algorithm for solving the singular linear systems is traversed. The convergence theorem of neural-network algorithm is presented and proved. The convergence theorem gives the theory to select the magnitude of learning ratio. In order to validate the validity of the algorithm introduced, the simulation example is presented. The simulation results show that it is significantly effective to solve the singular linear systems using neural-network algorithm.
出处
《湖南师范大学自然科学学报》
CAS
北大核心
2007年第3期38-41,共4页
Journal of Natural Science of Hunan Normal University
基金
湖南省软科学基金资助项目(2006ZK3028)
