Computing time-varying ML-weighted pseudoinverse by the Zhang neural networks. (English) Zbl 07310690
Summary: The Zhang neural network (ZNN), a recurrent neural network, proposed in 2001, is particularly effective in solving time-varying problems. It has shown high efficiency and excellent performance in various applications. The weighted pseudoinverse is a useful tool in solving and analyzing the constrained least-squares problems. In this paper, we propose a ZNN model for computing the weighted pseudoinverse of a time-varying matrix. We show that our model converges globally and exponentially to the solution and our system is robust at the presence of small errors. A Matlab Simulink implementation of our model is presented. Our convergence analysis is verified by our experiments on testing matrices. A comparison study shows that our model has superior performance over the conventional gradient-based neural networks.
MSC:
| 65F20 | Numerical solutions to overdetermined systems, pseudoinverses |
| 15A09 | Theory of matrix inversion and generalized inverses |
References:
| [1] | Cai, J.; Chen, G., The expression of \(####\) and its applications, Numer Math: J. Chinese Univ, 24, 4, 320-326 (2002) · Zbl 1025.15007 |
| [2] | Eldén, L., A weighted pseudoinverse, generalized singular values, and constrained least squares problems, BIT, 22, 4, 487-502 (1982) · Zbl 0509.65019 · doi:10.1007/BF01934412 |
| [3] | Galba, E. F.; Neka, V. S.; Sergienko, I. V., Weighted pseudoinverses and weighted normal pseudosolutions with singular weights, Comput. Math. And Math. Phys., 49, 8, 1281-1363 (2009) · Zbl 1199.65130 · doi:10.1134/S0965542509080016 |
| [4] | Liao, B.; Zhang, Y., Different complex ZFs leading to different complex ZNN models for time-varying complex generalized inverse matrices, IEEE Trans. Neural Netw. Learn Syst, 25, 9, 1621-1631 (2014) · doi:10.1109/TNNLS.2013.2271779 |
| [5] | Nashed, M. Z., The University of Wisconsin-Madison, 1976, Generalized inverses and applications, Proceedings of an Advanced Seminar Sponsored by the Mathematics Research Center, 8-10 (1973), New York, NY: Academic Press, Inc · Zbl 0324.00009 |
| [6] | Pian, J.; Chen, G., A unified approach for the recursive computation of weighted generalized inverses, J. Lanzhou Univ. (Nat. Sci.), 40, 6, 13-17 (2004) · Zbl 1069.65043 |
| [7] | Rao, C. R.; Mitra, S. K., Generalized Inverse of Matrices and Its Applications (1971), New York: John Wiley & Sons, New York · Zbl 0236.15004 |
| [8] | Rubini, L.; Cancelliere, R.; Gallinari, P.; Grosso, A.; Raiti, A.; Gennady, Agre; Pascal, Hitzler; Krisnadhi Adila, A.; Kuznetsov Sergei, O., Artificial Intelligence: Methodology, Systems, and Applications, Computational experience with pseudoinversion-based training of neural networks using random projection matrices (2014) |
| [9] | Wang, X.; Wei, Y.; Stanimirović, P. S., Complex neural network models for time-varying Drazin inverse, Neural Comput., 28, 12, 2790-2824 (2016) · Zbl 1474.68157 · doi:10.1162/NECO_a_00866 |
| [10] | Wei, M., Theory and Computation of Generalized Least Squares Problem (2006), Beijing: Science Press |
| [11] | Wei, M.; De Pierro, A. R., Upper perturbation bounds of weighted projections, weighted and constrained least squares problems, SIAM J. Matrix Anal. & Appl., 21, 3, 931-951 (2000) · Zbl 0953.65027 · doi:10.1137/S0895479898336306 |
| [12] | Wei, M.; Zhang, B., Structures and uniqueness conditions of MK-weighted pseudoinverses, BIT Numerical Mathematics, 34, 3, 437-450 (1994) · Zbl 0811.15003 · doi:10.1007/BF01935652 |
| [13] | Wei, Y., Recurrent neural networks for computing weighted Moore-Penrose inverse, Appl. Math. Comput., 116, 3, 279-287 (2000) · Zbl 1023.65030 · doi:10.1016/S0096-3003(99)00147-2 |
| [14] | Xiao, L., A finite-time convergent neural dynamics for online solution of time-varying linear complex matrix equation, Neurocomputing, 167, 1, 254-259 (2015) · doi:10.1016/j.neucom.2015.04.070 |
| [15] | Xiao, L., A nonlinearly activated neural dynamics and its finite-time solution to time-varying nonlinear equation, Neurocomputing, 173, 3, 1983-1988 (2016) · doi:10.1016/j.neucom.2015.08.031 |
| [16] | Xiao, L., A nonlinearly-activated neurodynamics model and its finite-time solution to equality constrained quadratic optimization with nonstationary coefficients, Appl. Soft Comput., 40, 252-259 (2016) · doi:10.1016/j.asoc.2015.11.023 |
| [17] | Xiao, L.; Liao, B., A convergence-accelerated Zhang neural network and its solution application to Lyapunov equation, Neurocomputing, 193, 213-218 (2016) · doi:10.1016/j.neucom.2016.02.021 |
| [18] | Xiao, L.; Lu, R., Finite-time solution to nonlinear equation using recurrent neural dynamics with a specially-constructed activation function, Neurocomputing, 151, 1, 246-251 (2015) · doi:10.1016/j.neucom.2014.09.047 |
| [19] | Zhang, Y., Revisit the analog computer and gradient-based neural system for matrix inversion, 1411-1416 (2005) |
| [20] | Zhang, Y.; Chen, K., Global exponential convergence and stability of Wang neural network for solving online linear equations, Electron. Lett., 44, 2, 145-146 (2008) · doi:10.1049/el:20081928 |
| [21] | Zhang, Y.; Chen, K.; Ma, W.; Li, X., MATLAB simulation of gradient-based neural network for online matrix inversion, 4682, 98-109 (2007) |
| [22] | Zhang, Y.; Ge, S., Design and analysis of a general recurrent neural network model for time-varying matrix inversion, IEEE Trans. Neural Netw., 16, 6, 1477-1490 (2005) · doi:10.1109/TNN.2005.857946 |
| [23] | Zhang, Y.; Jiang, D.; Wang, J., A recurrent neural network for solving Sylvester equation with time-varying coefficients, IEEE Trans Neural Netw, 13, 5, 1053-1063 (2002) · doi:10.1109/TNN.2002.1031938 |
| [24] | Zhang, Y.; Wang, J., Global exponential stability of recurrent neural networks for synthesizing linear feedback control systems via pole assignment, IEEE Trans Neural Netw, 13, 3, 633-644 (2002) · doi:10.1109/TNN.2002.1000129 |
| [25] | Zhang, Y.; Yang, Y.; Tan, N.; Cai, B., Zhang neural network solving for time-varying full-rank matrix Moore-Penrose inverse, Computing, 92, 2, 97-121 (2011) · Zbl 1226.93075 · doi:10.1007/s00607-010-0133-9 |
| [26] | Zhang, Y.; Yi, C., Zhang Neural Networks and Neural-Dynamic Method (2011), New York, NY: Nova Science Publishers, Inc |
| [27] | Zielke, G., Report on test matrices for generalized inverses, Computing, 36, 1-2, 105-162 (1986) · Zbl 0566.65026 · doi:10.1007/BF02238196 |
| [28] | Zivković, I. S.; Stanimirović, P. S.; Wei, Y., Recurrent neural network for computing outer inverse, Neural Comput., 28, 5, 970-998 (2016) · Zbl 1472.92050 · doi:10.1162/NECO_a_00821 |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
