用于奇异值分解的全并行神经网络
A NEW TOTAL PARALLEL NEURAL NETWORK FOR SVD
-
摘要: 本文提出一个实时奇异值分解(SVD)的全并行神经网络,给出并证明了它的有界性定理和稳定性定理,同时给出一个模拟例子。理论和模拟结果都说明所提出的神经网络对于SVD是有效的。Abstract: The paper presents a total parallell neural network for real-tune computation of SVD. Its boundedness and stability of the dynamical system are studied. Computer simulation is also given. The results of the theoretical analysis and simulation illustate that this neural network is feasible and efficient for SVD.
-
Golub G H, Van C F. Matrix Computations. Baltimore, MD: The Johns Hopkins University Press, 1989, 200-300.[2]Deprettere E F. ed, SVD and Signal Processing. North-Holland, Amsterden: 1988, 1-75.[3]Hopfield J J, Tank D W. Neural computation of decisions in optimization problems. Biol. Cybern, 1985, 52(2): 141-152.[4]Kennedy MP, Chua L O. Neural networks for nonlinear programming. IEEE Trans. on CAS, 1988, CAS-35(5): 554-562.[5]Oja E. A simplified neuron model as a principal component analyzer. J. Math. Biology. 1982, 15(1): 267-273.[6]Oja E. Principal components, minor components and linear neural networks[J].Neural Networks.1992, 5(6):927-935[7]Yuile A L, Kammen D M, Cohen D S. Quadrature and the development of orienation selective cortical cells by Hebb rules[J].Bio. Cybern.1989, 61(2):183-194[8]Cichocki A. Neural networks for singular value decompsition[J].Electron. Lett.1992, 28(8):784-786 -
计量
- 文章访问数: 2100
- HTML全文浏览量: 134
- PDF下载量: 389
- 被引次数: 0
下载:
