Abstract
Data modeling using a linear static model class leads to an unstructured low-rank approximation problem. When the approximation criterion is the Frobenius norm, the problem can be solved by the singular value decomposition. In general, however, unstructured low-rank approximation is a difficult nonconvex optimization problem. Section 4.1 presents a local optimization algorithm, based on alternating projections. An advantage of the alternating projections algorithm is that it is easily generalizable to missing data, nonnegativity, and other constraints on the data. approximate modeling problem leads to structured low-rank approximation. Section 4.2 presents a variable projection algorithm for affine structured low-rank approximation. This method is well suited for problems where one dimension of the matrix is small and the other one is large. This is the case in system identification, where the small dimension is determined by the model’s complexity and the large dimension is determined by the number of samples. The alternating projections and variable projection algorithms perform local optimization. An alternative approach for structured low-rank approximation is to solve a convex relaxation of the problem. A convex relaxation based on replacement of the rank constraint by a constraint on the nuclear norm is shown in Sect. 4.3. The nuclear norm heuristic is also applicable for matrix completion (missing data estimation). Section 4.4 generalize the variable projection method to solve low-rank matrix completion and approximation problems.
In general no simple relationships are satisfied exactly by the data. This discrepancy between observed data and simple relationships is often modelled by introducing stochastics. However, instead of stochastic uncertainty it is in our opinion primarily the complexity of reality which often prevents existence of simple exact models. In this case model errors do not reflect chance, but arise because a simple model can only give an approximate representation of complex systems.
Heij (1989)
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Absil PA, Mahony R, Sepulchre R (2008) Optimization algorithms on matrix manifolds. Princeton University Press, Princeton
Beck A, Ben-Tal A (2006) A global solution for the structured total least squares problem with block circulant matrices. SIAM J Matrix Anal Appl 27(1):238–255
Benner P, Mehrmann V, Sima V, Van Huffel S, Varga A (1999) SLICOT—a subroutine library in systems and control theory. Applied and computational control, signal and circuits, vol 1. Birkhauser, Boston, pp 499–539
Boumal N, Mishra B, Absil PA, Sepulchre R (2014) Manopt, a Matlab toolbox for optimization on manifolds. J Mach Learn Res 15:1455–1459. http://www.manopt.org
Chu M, Plemmons R (2003) Real-valued, low rank, circulant approximation. SIAM J Matrix Anal Appl 24(3):645–659
Dahl J, Vandenberghe L (2010) CVXOPT: Python software for convex optimization. http://abel.ee.ucla.edu/cvxopt
De Moor B (1993) Structured total least squares and \(L_2\) approximation problems. Linear Algebra Appl 188–189:163–207
Fazel M (2002) Matrix rank minimization with applications. PhD thesis, Electrical Engineering Department, Stanford University
Fazel M, Pong TK, Sun D, Tseng P (2013) Hankel matrix rank minimization with applications in system identification and realization. SIAM J Matrix Anal Appl 34(3):946–977
Galassi M et al (2017) GNU scientific library reference manual. http://www.gnu.org/software/gsl/
Grant M, Boyd S (2008) Graph implementations for nonsmooth convex programs. In: Blondel V, Boyd S, Kimura H (eds) Recent advances in learning and control. Springer, Berlin, pp 95–110. http://stanford.edu/~boyd/graph_dcp.html
Grant M, Boyd S (2017) CVX: Matlab software for disciplined convex programming. http://stanford.edu/~boyd/cvx
Heij C (1989) Deterministic identification of dynamical systems. Lecture notes in control and information sciences, vol 127. Springer, Berlin
Kailath T, Sayed A (1995) Displacement structure: theory and applications. SIAM Rev 37(3):297–386
Kailath T, Sayed A (1999) Fast reliable algorithms for matrices with structure. SIAM, Philadelphia
Keshavan R, Montanari A, Oh S (2010) Matrix completion from noisy entries. J Mach Learn Res 11:2057–2078
Kiers H (2002) Setting up alternating least squares and iterative majorization algorithms for solving various matrix optimization problems. Comput Stat Data Anal 41:157–170
Krijnen W (2006) Convergence of the sequence of parameters generated by alternating least squares algorithms. Comput Stat Data Anal 51:481–489
Liu Z, Vandenberghe L (2009) Interior-point method for nuclear norm approximation with application to system identification. SIAM J Matrix Anal Appl 31(3):1235–1256
Liu Z, Hansson A, Vandenberghe L (2013) Nuclear norm system identification with missing inputs and outputs. Control Lett 62:605–612
Manton J, Mahony R, Hua Y (2003) The geometry of weighted low-rank approximations. IEEE Trans Signal Process 51(2):500–514
Markovsky I (2012) How effective is the nuclear norm heuristic in solving data approximation problems? In: Proceedings of the 16th IFAC symposium on system identification, Brussels, pp 316–321
Markovsky I (2013) A software package for system identification in the behavioral setting. Control Eng Pract 21(10):1422–1436
Markovsky I, Usevich K (2013) Structured low-rank approximation with missing data. SIAM J Matrix Anal Appl 34(2):814–830
Markovsky I, Usevich K (2014) Software for weighted structured low-rank approximation. J Comput Appl Math 256:278–292
Markovsky I, Rastello M, Premoli A, Kukush A, Van Huffel S (2005) The element-wise weighted total least squares problem. Comput Stat Data Anal 50(1):181–209
Marquardt D (1963) An algorithm for least-squares estimation of nonlinear parameters. SIAM J Appl Math 11:431–441
Schuermans M, Markovsky I, Wentzell P, Van Huffel S (2005) On the equivalence between total least squares and maximum likelihood PCA. Anal Chim Acta 544(1–2):254–267
Usevich K, Markovsky I (2014) Variable projection for affinely structured low-rank approximation in weighted 2-norms. J Comput Appl Math 272:430–448
Wallin R, Hansson A (2014) Maximum likelihood estimation of linear SISO models subject to missing output data and missing input data. Int J Control 87(11):2354–2364
Wentzell P, Andrews D, Hamilton D, Faber K, Kowalski B (1997) Maximum likelihood principal component analysis. J Chemom 11:339–366
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2019 Springer International Publishing AG, part of Springer Nature
About this chapter
Cite this chapter
Markovsky, I. (2019). Approximate Modeling. In: Low-Rank Approximation. Communications and Control Engineering. Springer, Cham. https://doi.org/10.1007/978-3-319-89620-5_4
Download citation
DOI: https://doi.org/10.1007/978-3-319-89620-5_4
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-89619-9
Online ISBN: 978-3-319-89620-5
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)