Eigenvalues of C SMS (16) 1 A. Eigenvalues of A (32) 1 A (16,32) 1 A

Size: px
Start display at page:

Download "Eigenvalues of C SMS (16) 1 A. Eigenvalues of A (32) 1 A (16,32) 1 A"

Transcription

1 Negative Results for Multilevel Preconditioners in Image Deblurring C. R. Vogel Department of Mathematical Sciences Montana State University Bozeman, MT USA Abstract. A one-dimensional deconvolution problem is discretized and certain multilevel preconditioned iterative methods are applied to solve the resulting linear system. The numerical results suggest that multilevel multiplicative preconditioners may have no advantage over two-level multiplicative preconditioners. In fact, in the numerical experiments they perform worse than comparable two-level preconditioners. 1 Introduction Z In image deblurring, the goal is to estimate the true image u true from noisy, blurred data z(x) = k(x? y) u true (y) dy + (x): (1) Here represents noise in the recorded data, and the convolution kernel function k, which is called the point spread function in this application, is known. Since deconvolution is unstable with respect to perturbations in the data, regularization (i.e., stabilization which retains certain desired features of the true solution) must be applied. To a discretized version of the model equation (1), z = Ku true +, we apply standard (zero order) Tikhonov regularization, i.e., we minimize T (u) = jjku? zjj 2 + jjujj 2 ; > 0: (2) The is the regularization parameter. The resulting minimizer u solves the symmetric, positive denite (SPD) linear system Au = b; A = K K + I; (3) with b = K z. The superscript \" denotes matrix conjugate transpose. The system (3) is often quite large. For example, n = = 65; 536 unknowns arise from two-dimensional image data recorded on a pixel array. For real-time imaging applications, it is necessary to solve these systems very quickly. Due to these size and time constraints, iterative methods are required. Since the coecient matrix A in (3) is SPD, the conjugate gradient (CG) method is appropriate. A tends to be highly ill-conditioned, so preconditioning is needed to increase the convergence rate. (It should be noted that CG can

2 be applied to the unregularized system, and the iteration count becomes the regularization parameter. See for example [4] for details. We will not take this approach here.) Convolution integral operators typically lead to matrices K with Toeplitz structure (see [2]). Circulant preconditioners [7, 2, 3] have proven to be highly eective for large Toeplitz systems. Standard multigrid methods have also been implemented (see for example [1]), but their utility seems limited by the inability to nd a good \smoother", i.e., an iterative scheme which quickly damps high frequency components of solutions on ne grids. It is well-known that wavelet multilevel decompositions tend to \sparsify" the matrices which arise in the discretization of integral operators. The task which remains is to eciently solve the transformed system. Rieder [6] showed that fairly standard block iterative schemes (e.g., Jacobi and Gauss-Seidel iterations) are eective. (The correspondence between block index and grid level makes these methods \multilevel".) Hanke and Vogel [5, 8] extended Rieder's results to two-level preconditioners. Their analysis showed that multiplicative (i.e., Gauss- Seidel-like) preconditioners are generally far superior to additive (Jacobi-like) preconditioners in terms of convergence properties. In particular, they obtained the bounds on the condition numbers, cond(c add?1 1?2 as! 0; (4) cond(c mult?1 2?1 as! 0; (5) where C add and C mult denote the additive and multiplicative preconditioning matrices, respectively. They also presented numerical results indicating that these bounds were sharp. In addition to rapid convergence, these two-level schemes oer other advantages. Toeplitz structure is not required for ecient implementation. Essentially all that is needed is a means of computing matrix-vector products Av and a means of computing coarse-grid projections of vectors. See [8] for details. A disadvantage in certain situations is the need to invert the \coarse-grid representation" of the matrix A, which is the A 11 in equation (8) below. The constants b i in (4)-(5) depend on how well the integral operator is represented on the coarse grid. If is relatively small, then the b i 's also must be relatively small to maintain rapid convergence. This typically means that A 11 must be relatively large, and hence, expensive to invert. The cost of inverting relatively large coarse-grid representation matrices A 11 in the two-level case motivated our interest in multilevel schemes. We conducted a preliminary numerical study which suggested that, at least with obvious implementations, multilevel schemes oer no advantage over two-level schemes. In the nal section, we present the test problem used in this study. This is preceded by a brief sketch of multilevel iterative methods. 2 Multilevel Decomposition and Iterative Schemes Let the columns of V comprise the discrete Haar wavelet basis for R n, n = 2 p, normalized so that V V = I. Note that the discrete Haar wavelet vectors are

3 orthogonal with respect to the usual Euclidean inner product, so orthonormality can be achieved simply by rescaling these vectors. The system (3) can be transformed to ~Ax = ~ b; (6) where ~A = V AV; u = V x; ~ b = V b: (7) Partition the wavelet transformed matrix ~ A into blocks Aij, 1 i; j m, ~A = 2 64 A 11 A 12 : : : A 1m A 21 A 22 : : : A 2m..... A m1 A m2 : : : A mm 3 75 : (8) Multilevel iterative methods can be derived from a natural splitting of these blocks, i.e., ~A = L + D + U; (9) where U consists of the upper triangular blocks A ij ; j > i, D consists of the diagonal blocks A ii, and L = U. For instance, to derive a multilevel additive Jacobi iteration to solve (3), take an initial guess u 0, set x 0 = V u 0, iterate x +1 = D?1 ( ~ b? (L + U)x ); = 0; 1; : : : ; and then back-transform via (7) to obtain an approximate solution to the original system (3). To derive the additive Schwarz iteration presented by Rieder in [6], replace the A mm block in the block diagonal matrix D by I, where I is the identity matrix of the appropriate size. Similarly, one can derive multilevel Jacobi and additive Schwarz preconditioners. To apply such a Jacobi preconditioner to a vector r 2 R n, one rst applies the Haar wavelet transform to this vector, obtaining ~r = V r. One then computes x = D?1 ~r, and then back-transforms via (7) to get u = V D?1 V r = C?1 J r: The matrix C J = V DV is the multilevel Jacobi preconditioning matrix. To derive a multilevel additive Schwarz preconditioner, one again replaces the A mm block in D by I. These Jacobi/additive Schwarz iterative methods neglect o-diagonal terms in the block decomposition (8). Incorporating these o-diagonal terms leads to multiplicative iterative methods. Perhaps the simplest example is multilevel Gauss-Seidel iteration, which can be expressed as u = V x, where x is obtained from x +1 = (L + D)?1 ( ~ b? Ux ): (10) A multilevel multiplicative Schwarz iteration is obtained by again replacing the A mm block of D by I.

4 To obtain symmetric Gauss-Seidel/multiplicative Schwarz iterations, follow (10) by x +2 = (D + U)?1 ( ~ b? Lx +1 ): (11) To obtain the action of a multilevel symmetric Gauss-Seidel preconditioner on a vector r, replace ~ b in (10)-(11) by ~r = V r, set x = 0, and then back-transform via (7) to obtain C?1 SGS r = V (D + U)?1 (V r? L(L + D)?1 V r) = V (D + U)?1 D(L + D)?1 V r: (12) Consequently, the symmetric Gauss-Seidel preconditioning matrix is C SGS = V (L + D)D?1 (D + U)V : (13) Once again replacing the A mm block in D by I yields a corresponding multilevel symmetric multiplicative Schwarz preconditioner (SMS), which we denote by C SMS. 3 Numerical Results A symmetric Toeplitz matrix K = h toeplitz(k) was generated from a discretization k = (k(x 1 ); : : : ; k(x n )) of the Gaussian kernel function k(x) = exp(?x2 = 2 ) p ; 0 x 1; (14) 2 2 Here h = 1=n and x i = (i? 1)h; i = 1; : : : ; n. We selected the kernel width parameter = 0:05 and the number of grid points n = 2 7 = 128. The n n matrix K is extremely ill-conditioned, having eigenvalues which decay to zero like exp(? 2 j 2 ) for large j. The matrix A = K K + I was computed with regularization parameter = 10?3. The distribution of the eigenvalues of A is shown in the upper left subplot of Fig. 1. From this distribution, it can be seen that the eigencomponents corresponding to roughly the smallest 110 eigenvalues of K have been ltered out by the regularization. The value of the regularization parameter is nearly optimal for error-contaminated data whose signal-to-noise ratio is 100. We computed several two- and three-level SMS preconditioners for the system (3). The notation C SMS (r) denotes the two-level (m = 2 in equation (8)) SMS preconditioner with the \coarse grid" block A 11 of size r r. To obtain C SMS (r) the matrix A is transformed and partitioned into 2 2 blocks, cf. (8). The (n? r) (n? r) submatrix A 22 is replaced by I n?r, the splitting (9) is applied, and the right-hand-side of (13) is computed. The eigenvalues of the matrix products C SMS (r)?1 A were computed for coarse grid block sizes r = 16 and r = 32. The distributions of these eigenvalues are displayed in the upper right and lower left subplots of Fig. 1. The reduced relative spread and clustering of these eigenvalues ensures rapid CG convergence. Recall the eigenvalue

5 relative spread can be quantied by the condition number, which is the ratio of the largest to the smallest eigenvalue. With course grid block size r = 16, the condition number of C SMS (r)?1 A is 48:419, while for block size r = 32, the corresponding condition number is 1:4244. As one might expect, doubling the size of the coarse grid block A 11 substantially decreased the condition number. In contrast, the condition number of the matrix A (without preconditioning) is nearly Eigenvalues of A Eigenvalues of C SMS (16) 1 A Eigenvalues of C SMS (32) 1 A 10 0 Eigenvalues of C SMS (16,32) 1 A Fig. 1. Eigenvalue distributions for various multilevel symmetric multiplicative Schwarz preconditioned systems. Let C SMS (r; s) denote the three-level multiplicative Schwarz preconditioner whose coarse grid block A 11 in (8) has size rr and whose second diagonal block A 22 has size (s? r) (s? r). The third diagonal block A 33 is replaced by I n?s. With r = 16 and s = 32, the distribution of the eigenvalues of C SMS (r; s)?1 A is shown in the lower right subplot of Fig. 1. The corresponding condition number is 200:85. This is substantially worse than the result for the two-level preconditioner with coarse grid block size What is surprising is that this is worse than the result for the two-level preconditioner with coarse grid block size This comes in spite of the fact that much more work is required to apply C SMS (16; 32)?1 than to apply C SMS (16)?1. The condition numbers for the various matrix products C?1 A in the example presented above are summarized in column 2 (Test Case 1) of the table below. The I in column 1 indicates that no preconditioning is applied, i.e.,

6 the condition number of A appears across the corresponding row. In column 3 (Test Case 2), results are presented for the same kernel function, cf., equation (14), but the kernel width = :1 is increased, and the regularization parameter = 10?4 is decreased. As in Case 1, the two-level preconditioners outperform comparable three-level preconditioners. The dierence in performance is even more pronounced with the broader kernel than with the narrower kernel. In column 4 (Test Case 3), we present results for the sinc squared kernel, k(x) = (sin(x=)=(x=)) 2, with kernel width = :2, and regularization parameter = 10?5. The results are comparable to those obtained in Case 2. Preconditioner Test Case 1 Test Case 2 Test Case 3 C Condition No. of C?1 A Cond. No. C?1 A Cond. No. C?1 A I C SMS (16) C SMS (32) C SMS (16; 32) Conclusions. For all three test problems, the 3-level SMS preconditioners yielded larger condition numbers and less eigenvalue clustering than comparable 2-level SMS preconditioners. While these test cases may be unrealistically simple, they suggest that no advantage is to be gained by implementing multilevel preconditioners for more realistic (and complicated) problems. References 1. R. Chan, T. Chan, and W. L. Wan, Multigrid for Dierential-Convolution Problems Arising from Image Processing, in Proceedings of the Workshop on Sci. Comput., Springer Verlag, R. H. Chan and M. K. Ng, Conjugate Gradient Method for Toeplitz Systems, SIAM Review, 38 (1996), pp R. H. Chan, M. K. Ng and R. J. Plemmons, Generalization of Strang's preconditioner with applications to Toeplitz least squares problems, Numerical Lin. Alg. and Applications, 3 (1996), pp. 45{ H. Engl, M. Hanke, and A. Neubauer, Regularization of Inverse Problems, Kluwer Academic Publishers, Dordrecht, M. Hanke and C. R. Vogel, Two-Level Preconditioners for Regularized Inverse Problems I: Theory, Numer. Math., to appear. 6. A. Rieder, A wavelet multilevel method for ill-posed problems stablized by Tikhonov regularization, Numer. Math., 75 (1997), G. Strang, A Proposal for Toeplitz Matrix Calculations, Stud. Appl. Math, 74 (1986), C. R. Vogel and M. Hanke, Two-level preconditioners for regularized inverse problems II: Implementation and numerical results, SIAM J. Sci. Comput., to appear.

Computational Optical Imaging - Optique Numerique. -- Deconvolution --

Computational Optical Imaging - Optique Numerique. -- Deconvolution -- Computational Optical Imaging - Optique Numerique -- Deconvolution -- Winter 2014 Ivo Ihrke Deconvolution Ivo Ihrke Outline Deconvolution Theory example 1D deconvolution Fourier method Algebraic method

More information

Linköping University Electronic Press

Linköping University Electronic Press Linköping University Electronic Press Report A Preconditioned GMRES Method for Solving a 1D Sideways Heat Equation Zohreh Ranjbar and Lars Eldén LiTH-MAT-R, 348-296, No. 6 Available at: Linköping University

More information

Journal of Computational and Applied Mathematics

Journal of Computational and Applied Mathematics Journal of Computational and Applied Mathematics 226 (2009) 92 102 Contents lists available at ScienceDirect Journal of Computational and Applied Mathematics journal homepage: www.elsevier.com/locate/cam

More information

1. Introduction. Consider the computation of an approximate solution of the minimization problem

1. Introduction. Consider the computation of an approximate solution of the minimization problem A NEW TIKHONOV REGULARIZATION METHOD MARTIN FUHRY AND LOTHAR REICHEL Abstract. The numerical solution of linear discrete ill-posed problems typically requires regularization, i.e., replacement of the available

More information

Linear Algebra Review. Vectors

Linear Algebra Review. Vectors Linear Algebra Review By Tim K. Marks UCSD Borrows heavily from: Jana Kosecka kosecka@cs.gmu.edu http://cs.gmu.edu/~kosecka/cs682.html Virginia de Sa Cogsci 8F Linear Algebra review UCSD Vectors The length

More information

7 Gaussian Elimination and LU Factorization

7 Gaussian Elimination and LU Factorization 7 Gaussian Elimination and LU Factorization In this final section on matrix factorization methods for solving Ax = b we want to take a closer look at Gaussian elimination (probably the best known method

More information

MATRIX ALGEBRA AND SYSTEMS OF EQUATIONS

MATRIX ALGEBRA AND SYSTEMS OF EQUATIONS MATRIX ALGEBRA AND SYSTEMS OF EQUATIONS Systems of Equations and Matrices Representation of a linear system The general system of m equations in n unknowns can be written a x + a 2 x 2 + + a n x n b a

More information

MATRIX ALGEBRA AND SYSTEMS OF EQUATIONS. + + x 2. x n. a 11 a 12 a 1n b 1 a 21 a 22 a 2n b 2 a 31 a 32 a 3n b 3. a m1 a m2 a mn b m

MATRIX ALGEBRA AND SYSTEMS OF EQUATIONS. + + x 2. x n. a 11 a 12 a 1n b 1 a 21 a 22 a 2n b 2 a 31 a 32 a 3n b 3. a m1 a m2 a mn b m MATRIX ALGEBRA AND SYSTEMS OF EQUATIONS 1. SYSTEMS OF EQUATIONS AND MATRICES 1.1. Representation of a linear system. The general system of m equations in n unknowns can be written a 11 x 1 + a 12 x 2 +

More information

13 MATH FACTS 101. 2 a = 1. 7. The elements of a vector have a graphical interpretation, which is particularly easy to see in two or three dimensions.

13 MATH FACTS 101. 2 a = 1. 7. The elements of a vector have a graphical interpretation, which is particularly easy to see in two or three dimensions. 3 MATH FACTS 0 3 MATH FACTS 3. Vectors 3.. Definition We use the overhead arrow to denote a column vector, i.e., a linear segment with a direction. For example, in three-space, we write a vector in terms

More information

Numerical Methods I Solving Linear Systems: Sparse Matrices, Iterative Methods and Non-Square Systems

Numerical Methods I Solving Linear Systems: Sparse Matrices, Iterative Methods and Non-Square Systems Numerical Methods I Solving Linear Systems: Sparse Matrices, Iterative Methods and Non-Square Systems Aleksandar Donev Courant Institute, NYU 1 donev@courant.nyu.edu 1 Course G63.2010.001 / G22.2420-001,

More information

SOLVING LINEAR SYSTEMS

SOLVING LINEAR SYSTEMS SOLVING LINEAR SYSTEMS Linear systems Ax = b occur widely in applied mathematics They occur as direct formulations of real world problems; but more often, they occur as a part of the numerical analysis

More information

The Singular Value Decomposition in Symmetric (Löwdin) Orthogonalization and Data Compression

The Singular Value Decomposition in Symmetric (Löwdin) Orthogonalization and Data Compression The Singular Value Decomposition in Symmetric (Löwdin) Orthogonalization and Data Compression The SVD is the most generally applicable of the orthogonal-diagonal-orthogonal type matrix decompositions Every

More information

15.062 Data Mining: Algorithms and Applications Matrix Math Review

15.062 Data Mining: Algorithms and Applications Matrix Math Review .6 Data Mining: Algorithms and Applications Matrix Math Review The purpose of this document is to give a brief review of selected linear algebra concepts that will be useful for the course and to develop

More information

Similar matrices and Jordan form

Similar matrices and Jordan form Similar matrices and Jordan form We ve nearly covered the entire heart of linear algebra once we ve finished singular value decompositions we ll have seen all the most central topics. A T A is positive

More information

October 3rd, 2012. Linear Algebra & Properties of the Covariance Matrix

October 3rd, 2012. Linear Algebra & Properties of the Covariance Matrix Linear Algebra & Properties of the Covariance Matrix October 3rd, 2012 Estimation of r and C Let rn 1, rn, t..., rn T be the historical return rates on the n th asset. rn 1 rṇ 2 r n =. r T n n = 1, 2,...,

More information

Solving linear equations on parallel distributed memory architectures by extrapolation Christer Andersson Abstract Extrapolation methods can be used to accelerate the convergence of vector sequences. It

More information

Automatic parameter setting for Arnoldi-Tikhonov methods

Automatic parameter setting for Arnoldi-Tikhonov methods Automatic parameter setting for Arnoldi-Tikhonov methods S. Gazzola, P. Novati Department of Mathematics University of Padova, Italy March 16, 2012 Abstract In the framework of iterative regularization

More information

Lecture 5: Singular Value Decomposition SVD (1)

Lecture 5: Singular Value Decomposition SVD (1) EEM3L1: Numerical and Analytical Techniques Lecture 5: Singular Value Decomposition SVD (1) EE3L1, slide 1, Version 4: 25-Sep-02 Motivation for SVD (1) SVD = Singular Value Decomposition Consider the system

More information

Chapter 6. Orthogonality

Chapter 6. Orthogonality 6.3 Orthogonal Matrices 1 Chapter 6. Orthogonality 6.3 Orthogonal Matrices Definition 6.4. An n n matrix A is orthogonal if A T A = I. Note. We will see that the columns of an orthogonal matrix must be

More information

Multigrid preconditioning for nonlinear (degenerate) parabolic equations with application to monument degradation

Multigrid preconditioning for nonlinear (degenerate) parabolic equations with application to monument degradation Multigrid preconditioning for nonlinear (degenerate) parabolic equations with application to monument degradation M. Donatelli 1 M. Semplice S. Serra-Capizzano 1 1 Department of Science and High Technology

More information

5. Orthogonal matrices

5. Orthogonal matrices L Vandenberghe EE133A (Spring 2016) 5 Orthogonal matrices matrices with orthonormal columns orthogonal matrices tall matrices with orthonormal columns complex matrices with orthonormal columns 5-1 Orthonormal

More information

Inner Product Spaces and Orthogonality

Inner Product Spaces and Orthogonality Inner Product Spaces and Orthogonality week 3-4 Fall 2006 Dot product of R n The inner product or dot product of R n is a function, defined by u, v a b + a 2 b 2 + + a n b n for u a, a 2,, a n T, v b,

More information

Abstract: We describe the beautiful LU factorization of a square matrix (or how to write Gaussian elimination in terms of matrix multiplication).

Abstract: We describe the beautiful LU factorization of a square matrix (or how to write Gaussian elimination in terms of matrix multiplication). MAT 2 (Badger, Spring 202) LU Factorization Selected Notes September 2, 202 Abstract: We describe the beautiful LU factorization of a square matrix (or how to write Gaussian elimination in terms of matrix

More information

[1] Diagonal factorization

[1] Diagonal factorization 8.03 LA.6: Diagonalization and Orthogonal Matrices [ Diagonal factorization [2 Solving systems of first order differential equations [3 Symmetric and Orthonormal Matrices [ Diagonal factorization Recall:

More information

Numerical Methods For Image Restoration

Numerical Methods For Image Restoration Numerical Methods For Image Restoration CIRAM Alessandro Lanza University of Bologna, Italy Faculty of Engineering CIRAM Outline 1. Image Restoration as an inverse problem 2. Image degradation models:

More information

Yousef Saad University of Minnesota Computer Science and Engineering. CRM Montreal - April 30, 2008

Yousef Saad University of Minnesota Computer Science and Engineering. CRM Montreal - April 30, 2008 A tutorial on: Iterative methods for Sparse Matrix Problems Yousef Saad University of Minnesota Computer Science and Engineering CRM Montreal - April 30, 2008 Outline Part 1 Sparse matrices and sparsity

More information

Chapter 12 Modal Decomposition of State-Space Models 12.1 Introduction The solutions obtained in previous chapters, whether in time domain or transfor

Chapter 12 Modal Decomposition of State-Space Models 12.1 Introduction The solutions obtained in previous chapters, whether in time domain or transfor Lectures on Dynamic Systems and Control Mohammed Dahleh Munther A. Dahleh George Verghese Department of Electrical Engineering and Computer Science Massachuasetts Institute of Technology 1 1 c Chapter

More information

6. Cholesky factorization

6. Cholesky factorization 6. Cholesky factorization EE103 (Fall 2011-12) triangular matrices forward and backward substitution the Cholesky factorization solving Ax = b with A positive definite inverse of a positive definite matrix

More information

Introduction to Matrix Algebra

Introduction to Matrix Algebra Psychology 7291: Multivariate Statistics (Carey) 8/27/98 Matrix Algebra - 1 Introduction to Matrix Algebra Definitions: A matrix is a collection of numbers ordered by rows and columns. It is customary

More information

Solving Linear Systems, Continued and The Inverse of a Matrix

Solving Linear Systems, Continued and The Inverse of a Matrix , Continued and The of a Matrix Calculus III Summer 2013, Session II Monday, July 15, 2013 Agenda 1. The rank of a matrix 2. The inverse of a square matrix Gaussian Gaussian solves a linear system by reducing

More information

By choosing to view this document, you agree to all provisions of the copyright laws protecting it.

By choosing to view this document, you agree to all provisions of the copyright laws protecting it. This material is posted here with permission of the IEEE Such permission of the IEEE does not in any way imply IEEE endorsement of any of Helsinki University of Technology's products or services Internal

More information

P164 Tomographic Velocity Model Building Using Iterative Eigendecomposition

P164 Tomographic Velocity Model Building Using Iterative Eigendecomposition P164 Tomographic Velocity Model Building Using Iterative Eigendecomposition K. Osypov* (WesternGeco), D. Nichols (WesternGeco), M. Woodward (WesternGeco) & C.E. Yarman (WesternGeco) SUMMARY Tomographic

More information

Dynamic Eigenvalues for Scalar Linear Time-Varying Systems

Dynamic Eigenvalues for Scalar Linear Time-Varying Systems Dynamic Eigenvalues for Scalar Linear Time-Varying Systems P. van der Kloet and F.L. Neerhoff Department of Electrical Engineering Delft University of Technology Mekelweg 4 2628 CD Delft The Netherlands

More information

Similarity and Diagonalization. Similar Matrices

Similarity and Diagonalization. Similar Matrices MATH022 Linear Algebra Brief lecture notes 48 Similarity and Diagonalization Similar Matrices Let A and B be n n matrices. We say that A is similar to B if there is an invertible n n matrix P such that

More information

Section 6.1 - Inner Products and Norms

Section 6.1 - Inner Products and Norms Section 6.1 - Inner Products and Norms Definition. Let V be a vector space over F {R, C}. An inner product on V is a function that assigns, to every ordered pair of vectors x and y in V, a scalar in F,

More information

Factorization Theorems

Factorization Theorems Chapter 7 Factorization Theorems This chapter highlights a few of the many factorization theorems for matrices While some factorization results are relatively direct, others are iterative While some factorization

More information

The Characteristic Polynomial

The Characteristic Polynomial Physics 116A Winter 2011 The Characteristic Polynomial 1 Coefficients of the characteristic polynomial Consider the eigenvalue problem for an n n matrix A, A v = λ v, v 0 (1) The solution to this problem

More information

Linear Algebra Methods for Data Mining

Linear Algebra Methods for Data Mining Linear Algebra Methods for Data Mining Saara Hyvönen, Saara.Hyvonen@cs.helsinki.fi Spring 2007 Lecture 3: QR, least squares, linear regression Linear Algebra Methods for Data Mining, Spring 2007, University

More information

Inner Product Spaces

Inner Product Spaces Math 571 Inner Product Spaces 1. Preliminaries An inner product space is a vector space V along with a function, called an inner product which associates each pair of vectors u, v with a scalar u, v, and

More information

1 VECTOR SPACES AND SUBSPACES

1 VECTOR SPACES AND SUBSPACES 1 VECTOR SPACES AND SUBSPACES What is a vector? Many are familiar with the concept of a vector as: Something which has magnitude and direction. an ordered pair or triple. a description for quantities such

More information

DATA ANALYSIS II. Matrix Algorithms

DATA ANALYSIS II. Matrix Algorithms DATA ANALYSIS II Matrix Algorithms Similarity Matrix Given a dataset D = {x i }, i=1,..,n consisting of n points in R d, let A denote the n n symmetric similarity matrix between the points, given as where

More information

Matrix Differentiation

Matrix Differentiation 1 Introduction Matrix Differentiation ( and some other stuff ) Randal J. Barnes Department of Civil Engineering, University of Minnesota Minneapolis, Minnesota, USA Throughout this presentation I have

More information

Numerical Methods I Eigenvalue Problems

Numerical Methods I Eigenvalue Problems Numerical Methods I Eigenvalue Problems Aleksandar Donev Courant Institute, NYU 1 donev@courant.nyu.edu 1 Course G63.2010.001 / G22.2420-001, Fall 2010 September 30th, 2010 A. Donev (Courant Institute)

More information

Linear Algebra and TI 89

Linear Algebra and TI 89 Linear Algebra and TI 89 Abdul Hassen and Jay Schiffman This short manual is a quick guide to the use of TI89 for Linear Algebra. We do this in two sections. In the first section, we will go over the editing

More information

Linear Algebra: Determinants, Inverses, Rank

Linear Algebra: Determinants, Inverses, Rank D Linear Algebra: Determinants, Inverses, Rank D 1 Appendix D: LINEAR ALGEBRA: DETERMINANTS, INVERSES, RANK TABLE OF CONTENTS Page D.1. Introduction D 3 D.2. Determinants D 3 D.2.1. Some Properties of

More information

CITY UNIVERSITY LONDON. BEng Degree in Computer Systems Engineering Part II BSc Degree in Computer Systems Engineering Part III PART 2 EXAMINATION

CITY UNIVERSITY LONDON. BEng Degree in Computer Systems Engineering Part II BSc Degree in Computer Systems Engineering Part III PART 2 EXAMINATION No: CITY UNIVERSITY LONDON BEng Degree in Computer Systems Engineering Part II BSc Degree in Computer Systems Engineering Part III PART 2 EXAMINATION ENGINEERING MATHEMATICS 2 (resit) EX2005 Date: August

More information

10.2 ITERATIVE METHODS FOR SOLVING LINEAR SYSTEMS. The Jacobi Method

10.2 ITERATIVE METHODS FOR SOLVING LINEAR SYSTEMS. The Jacobi Method 578 CHAPTER 1 NUMERICAL METHODS 1. ITERATIVE METHODS FOR SOLVING LINEAR SYSTEMS As a numerical technique, Gaussian elimination is rather unusual because it is direct. That is, a solution is obtained after

More information

Sparse recovery and compressed sensing in inverse problems

Sparse recovery and compressed sensing in inverse problems Gerd Teschke (7. Juni 2010) 1/68 Sparse recovery and compressed sensing in inverse problems Gerd Teschke (joint work with Evelyn Herrholz) Institute for Computational Mathematics in Science and Technology

More information

MATH 423 Linear Algebra II Lecture 38: Generalized eigenvectors. Jordan canonical form (continued).

MATH 423 Linear Algebra II Lecture 38: Generalized eigenvectors. Jordan canonical form (continued). MATH 423 Linear Algebra II Lecture 38: Generalized eigenvectors Jordan canonical form (continued) Jordan canonical form A Jordan block is a square matrix of the form λ 1 0 0 0 0 λ 1 0 0 0 0 λ 0 0 J = 0

More information

α = u v. In other words, Orthogonal Projection

α = u v. In other words, Orthogonal Projection Orthogonal Projection Given any nonzero vector v, it is possible to decompose an arbitrary vector u into a component that points in the direction of v and one that points in a direction orthogonal to v

More information

Correlation and Convolution Class Notes for CMSC 426, Fall 2005 David Jacobs

Correlation and Convolution Class Notes for CMSC 426, Fall 2005 David Jacobs Correlation and Convolution Class otes for CMSC 46, Fall 5 David Jacobs Introduction Correlation and Convolution are basic operations that we will perform to extract information from images. They are in

More information

Lecture 1: Schur s Unitary Triangularization Theorem

Lecture 1: Schur s Unitary Triangularization Theorem Lecture 1: Schur s Unitary Triangularization Theorem This lecture introduces the notion of unitary equivalence and presents Schur s theorem and some of its consequences It roughly corresponds to Sections

More information

Math 115A HW4 Solutions University of California, Los Angeles. 5 2i 6 + 4i. (5 2i)7i (6 + 4i)( 3 + i) = 35i + 14 ( 22 6i) = 36 + 41i.

Math 115A HW4 Solutions University of California, Los Angeles. 5 2i 6 + 4i. (5 2i)7i (6 + 4i)( 3 + i) = 35i + 14 ( 22 6i) = 36 + 41i. Math 5A HW4 Solutions September 5, 202 University of California, Los Angeles Problem 4..3b Calculate the determinant, 5 2i 6 + 4i 3 + i 7i Solution: The textbook s instructions give us, (5 2i)7i (6 + 4i)(

More information

Vector and Matrix Norms

Vector and Matrix Norms Chapter 1 Vector and Matrix Norms 11 Vector Spaces Let F be a field (such as the real numbers, R, or complex numbers, C) with elements called scalars A Vector Space, V, over the field F is a non-empty

More information

Using row reduction to calculate the inverse and the determinant of a square matrix

Using row reduction to calculate the inverse and the determinant of a square matrix Using row reduction to calculate the inverse and the determinant of a square matrix Notes for MATH 0290 Honors by Prof. Anna Vainchtein 1 Inverse of a square matrix An n n square matrix A is called invertible

More information

Solving Systems of Linear Equations

Solving Systems of Linear Equations LECTURE 5 Solving Systems of Linear Equations Recall that we introduced the notion of matrices as a way of standardizing the expression of systems of linear equations In today s lecture I shall show how

More information

MATH10212 Linear Algebra. Systems of Linear Equations. Definition. An n-dimensional vector is a row or a column of n numbers (or letters): a 1.

MATH10212 Linear Algebra. Systems of Linear Equations. Definition. An n-dimensional vector is a row or a column of n numbers (or letters): a 1. MATH10212 Linear Algebra Textbook: D. Poole, Linear Algebra: A Modern Introduction. Thompson, 2006. ISBN 0-534-40596-7. Systems of Linear Equations Definition. An n-dimensional vector is a row or a column

More information

MATH 551 - APPLIED MATRIX THEORY

MATH 551 - APPLIED MATRIX THEORY MATH 55 - APPLIED MATRIX THEORY FINAL TEST: SAMPLE with SOLUTIONS (25 points NAME: PROBLEM (3 points A web of 5 pages is described by a directed graph whose matrix is given by A Do the following ( points

More information

The Image Deblurring Problem

The Image Deblurring Problem page 1 Chapter 1 The Image Deblurring Problem You cannot depend on your eyes when your imagination is out of focus. Mark Twain When we use a camera, we want the recorded image to be a faithful representation

More information

8 Square matrices continued: Determinants

8 Square matrices continued: Determinants 8 Square matrices continued: Determinants 8. Introduction Determinants give us important information about square matrices, and, as we ll soon see, are essential for the computation of eigenvalues. You

More information

Operation Count; Numerical Linear Algebra

Operation Count; Numerical Linear Algebra 10 Operation Count; Numerical Linear Algebra 10.1 Introduction Many computations are limited simply by the sheer number of required additions, multiplications, or function evaluations. If floating-point

More information

160 CHAPTER 4. VECTOR SPACES

160 CHAPTER 4. VECTOR SPACES 160 CHAPTER 4. VECTOR SPACES 4. Rank and Nullity In this section, we look at relationships between the row space, column space, null space of a matrix and its transpose. We will derive fundamental results

More information

Iterative Methods for Solving Linear Systems

Iterative Methods for Solving Linear Systems Chapter 5 Iterative Methods for Solving Linear Systems 5.1 Convergence of Sequences of Vectors and Matrices In Chapter 2 we have discussed some of the main methods for solving systems of linear equations.

More information

Nonlinear Iterative Partial Least Squares Method

Nonlinear Iterative Partial Least Squares Method Numerical Methods for Determining Principal Component Analysis Abstract Factors Béchu, S., Richard-Plouet, M., Fernandez, V., Walton, J., and Fairley, N. (2016) Developments in numerical treatments for

More information

x1 x 2 x 3 y 1 y 2 y 3 x 1 y 2 x 2 y 1 0.

x1 x 2 x 3 y 1 y 2 y 3 x 1 y 2 x 2 y 1 0. Cross product 1 Chapter 7 Cross product We are getting ready to study integration in several variables. Until now we have been doing only differential calculus. One outcome of this study will be our ability

More information

Recall the basic property of the transpose (for any A): v A t Aw = v w, v, w R n.

Recall the basic property of the transpose (for any A): v A t Aw = v w, v, w R n. ORTHOGONAL MATRICES Informally, an orthogonal n n matrix is the n-dimensional analogue of the rotation matrices R θ in R 2. When does a linear transformation of R 3 (or R n ) deserve to be called a rotation?

More information

6.3 Multigrid Methods

6.3 Multigrid Methods 6.3 Multigrid Methods The Jacobi and Gauss-Seidel iterations produce smooth errors. The error vector e has its high frequencies nearly removed in a few iterations. But low frequencies are reduced very

More information

CS3220 Lecture Notes: QR factorization and orthogonal transformations

CS3220 Lecture Notes: QR factorization and orthogonal transformations CS3220 Lecture Notes: QR factorization and orthogonal transformations Steve Marschner Cornell University 11 March 2009 In this lecture I ll talk about orthogonal matrices and their properties, discuss

More information

More than you wanted to know about quadratic forms

More than you wanted to know about quadratic forms CALIFORNIA INSTITUTE OF TECHNOLOGY Division of the Humanities and Social Sciences More than you wanted to know about quadratic forms KC Border Contents 1 Quadratic forms 1 1.1 Quadratic forms on the unit

More information

TESLA Report 2003-03

TESLA Report 2003-03 TESLA Report 23-3 A multigrid based 3D space-charge routine in the tracking code GPT Gisela Pöplau, Ursula van Rienen, Marieke de Loos and Bas van der Geer Institute of General Electrical Engineering,

More information

Automatic parameter setting for Arnoldi-Tikhonov methods

Automatic parameter setting for Arnoldi-Tikhonov methods Automatic parameter setting for Arnoldi-Tikhonov methods S. Gazzola, P. Novati Department of Mathematics University of Padova, Italy Abstract In the framework of iterative regularization techniques for

More information

Lecture 5 Least-squares

Lecture 5 Least-squares EE263 Autumn 2007-08 Stephen Boyd Lecture 5 Least-squares least-squares (approximate) solution of overdetermined equations projection and orthogonality principle least-squares estimation BLUE property

More information

1 2 3 1 1 2 x = + x 2 + x 4 1 0 1

1 2 3 1 1 2 x = + x 2 + x 4 1 0 1 (d) If the vector b is the sum of the four columns of A, write down the complete solution to Ax = b. 1 2 3 1 1 2 x = + x 2 + x 4 1 0 0 1 0 1 2. (11 points) This problem finds the curve y = C + D 2 t which

More information

1 Determinants and the Solvability of Linear Systems

1 Determinants and the Solvability of Linear Systems 1 Determinants and the Solvability of Linear Systems In the last section we learned how to use Gaussian elimination to solve linear systems of n equations in n unknowns The section completely side-stepped

More information

1 Review of Least Squares Solutions to Overdetermined Systems

1 Review of Least Squares Solutions to Overdetermined Systems cs4: introduction to numerical analysis /9/0 Lecture 7: Rectangular Systems and Numerical Integration Instructor: Professor Amos Ron Scribes: Mark Cowlishaw, Nathanael Fillmore Review of Least Squares

More information

Continuity of the Perron Root

Continuity of the Perron Root Linear and Multilinear Algebra http://dx.doi.org/10.1080/03081087.2014.934233 ArXiv: 1407.7564 (http://arxiv.org/abs/1407.7564) Continuity of the Perron Root Carl D. Meyer Department of Mathematics, North

More information

Recovery of missing segments and lines in images. H. G. Feichtinger and T. Strohmer. University of Vienna, Department of Mathematics

Recovery of missing segments and lines in images. H. G. Feichtinger and T. Strohmer. University of Vienna, Department of Mathematics Recovery of missing segments and lines in images H. G. Feichtinger and T. Strohmer NUGHAG { Numerical Harmonic Analysis Group University of Vienna, Department of Mathematics Strudlhofgasse 4, A-1090 Vienna,

More information

ALGEBRAIC EIGENVALUE PROBLEM

ALGEBRAIC EIGENVALUE PROBLEM ALGEBRAIC EIGENVALUE PROBLEM BY J. H. WILKINSON, M.A. (Cantab.), Sc.D. Technische Universes! Dsrmstedt FACHBEREICH (NFORMATiK BIBL1OTHEK Sachgebieto:. Standort: CLARENDON PRESS OXFORD 1965 Contents 1.

More information

Unit 18 Determinants

Unit 18 Determinants Unit 18 Determinants Every square matrix has a number associated with it, called its determinant. In this section, we determine how to calculate this number, and also look at some of the properties of

More information

Applications to Data Smoothing and Image Processing I

Applications to Data Smoothing and Image Processing I Applications to Data Smoothing and Image Processing I MA 348 Kurt Bryan Signals and Images Let t denote time and consider a signal a(t) on some time interval, say t. We ll assume that the signal a(t) is

More information

3 Orthogonal Vectors and Matrices

3 Orthogonal Vectors and Matrices 3 Orthogonal Vectors and Matrices The linear algebra portion of this course focuses on three matrix factorizations: QR factorization, singular valued decomposition (SVD), and LU factorization The first

More information

Understanding and Applying Kalman Filtering

Understanding and Applying Kalman Filtering Understanding and Applying Kalman Filtering Lindsay Kleeman Department of Electrical and Computer Systems Engineering Monash University, Clayton 1 Introduction Objectives: 1. Provide a basic understanding

More information

Inner product. Definition of inner product

Inner product. Definition of inner product Math 20F Linear Algebra Lecture 25 1 Inner product Review: Definition of inner product. Slide 1 Norm and distance. Orthogonal vectors. Orthogonal complement. Orthogonal basis. Definition of inner product

More information

Blind Deconvolution of Barcodes via Dictionary Analysis and Wiener Filter of Barcode Subsections

Blind Deconvolution of Barcodes via Dictionary Analysis and Wiener Filter of Barcode Subsections Blind Deconvolution of Barcodes via Dictionary Analysis and Wiener Filter of Barcode Subsections Maximilian Hung, Bohyun B. Kim, Xiling Zhang August 17, 2013 Abstract While current systems already provide

More information

Continued Fractions and the Euclidean Algorithm

Continued Fractions and the Euclidean Algorithm Continued Fractions and the Euclidean Algorithm Lecture notes prepared for MATH 326, Spring 997 Department of Mathematics and Statistics University at Albany William F Hammond Table of Contents Introduction

More information

General Framework for an Iterative Solution of Ax b. Jacobi s Method

General Framework for an Iterative Solution of Ax b. Jacobi s Method 2.6 Iterative Solutions of Linear Systems 143 2.6 Iterative Solutions of Linear Systems Consistent linear systems in real life are solved in one of two ways: by direct calculation (using a matrix factorization,

More information

Orthogonal Diagonalization of Symmetric Matrices

Orthogonal Diagonalization of Symmetric Matrices MATH10212 Linear Algebra Brief lecture notes 57 Gram Schmidt Process enables us to find an orthogonal basis of a subspace. Let u 1,..., u k be a basis of a subspace V of R n. We begin the process of finding

More information

December 4, 2013 MATH 171 BASIC LINEAR ALGEBRA B. KITCHENS

December 4, 2013 MATH 171 BASIC LINEAR ALGEBRA B. KITCHENS December 4, 2013 MATH 171 BASIC LINEAR ALGEBRA B KITCHENS The equation 1 Lines in two-dimensional space (1) 2x y = 3 describes a line in two-dimensional space The coefficients of x and y in the equation

More information

Lecture notes on linear algebra

Lecture notes on linear algebra Lecture notes on linear algebra David Lerner Department of Mathematics University of Kansas These are notes of a course given in Fall, 2007 and 2008 to the Honors sections of our elementary linear algebra

More information

LINEAR ALGEBRA. September 23, 2010

LINEAR ALGEBRA. September 23, 2010 LINEAR ALGEBRA September 3, 00 Contents 0. LU-decomposition.................................... 0. Inverses and Transposes................................. 0.3 Column Spaces and NullSpaces.............................

More information

Au = = = 3u. Aw = = = 2w. so the action of A on u and w is very easy to picture: it simply amounts to a stretching by 3 and 2, respectively.

Au = = = 3u. Aw = = = 2w. so the action of A on u and w is very easy to picture: it simply amounts to a stretching by 3 and 2, respectively. Chapter 7 Eigenvalues and Eigenvectors In this last chapter of our exploration of Linear Algebra we will revisit eigenvalues and eigenvectors of matrices, concepts that were already introduced in Geometry

More information

Eigenvalues, Eigenvectors, Matrix Factoring, and Principal Components

Eigenvalues, Eigenvectors, Matrix Factoring, and Principal Components Eigenvalues, Eigenvectors, Matrix Factoring, and Principal Components The eigenvalues and eigenvectors of a square matrix play a key role in some important operations in statistics. In particular, they

More information

Statistical Machine Learning

Statistical Machine Learning Statistical Machine Learning UoC Stats 37700, Winter quarter Lecture 4: classical linear and quadratic discriminants. 1 / 25 Linear separation For two classes in R d : simple idea: separate the classes

More information

The Basics of FEA Procedure

The Basics of FEA Procedure CHAPTER 2 The Basics of FEA Procedure 2.1 Introduction This chapter discusses the spring element, especially for the purpose of introducing various concepts involved in use of the FEA technique. A spring

More information

Systems of Linear Equations

Systems of Linear Equations Systems of Linear Equations Beifang Chen Systems of linear equations Linear systems A linear equation in variables x, x,, x n is an equation of the form a x + a x + + a n x n = b, where a, a,, a n and

More information

Rotation Matrices and Homogeneous Transformations

Rotation Matrices and Homogeneous Transformations Rotation Matrices and Homogeneous Transformations A coordinate frame in an n-dimensional space is defined by n mutually orthogonal unit vectors. In particular, for a two-dimensional (2D) space, i.e., n

More information

Numerical Analysis Lecture Notes

Numerical Analysis Lecture Notes Numerical Analysis Lecture Notes Peter J. Olver 6. Eigenvalues and Singular Values In this section, we collect together the basic facts about eigenvalues and eigenvectors. From a geometrical viewpoint,

More information

Linköping University Electronic Press

Linköping University Electronic Press Linköping University Electronic Press Report Well-posed boundary conditions for the shallow water equations Sarmad Ghader and Jan Nordström Series: LiTH-MAT-R, 0348-960, No. 4 Available at: Linköping University

More information

Solving simultaneous equations using the inverse matrix

Solving simultaneous equations using the inverse matrix Solving simultaneous equations using the inverse matrix 8.2 Introduction The power of matrix algebra is seen in the representation of a system of simultaneous linear equations as a matrix equation. Matrix

More information

Notes on Orthogonal and Symmetric Matrices MENU, Winter 2013

Notes on Orthogonal and Symmetric Matrices MENU, Winter 2013 Notes on Orthogonal and Symmetric Matrices MENU, Winter 201 These notes summarize the main properties and uses of orthogonal and symmetric matrices. We covered quite a bit of material regarding these topics,

More information