Course syllabus
Course PM
This page contains the program of the course: lectures, exercise sessions and computer labs. Other information, such as learning outcomes, teachers, literature and examination, are in a separate course PM.
Registering for courses for PhD students
Program
Zoom-link to the home examination at 24.08.2020, 8:30-12.30 (password: 994244)
The schedule of the course is in TimeEdit.
Lectures
Day | Time | Place | Remarks |
---|---|---|---|
MON | 13:15-15:00 | Euler | Lecture |
WED | 13:15-15:00 | MVF24, MVF25 | Computer exercises |
THU | 10:00-11:45 | Pascal | Lecture |
FRI | 13:15-15:00 | MVF24, MVF25 | Computer exercises |
29.10.2019 | 14.00-18.00 | SB | Examination |
09.01.20129 | 14.00-18.00 | SB | Examination |
Grades at Examination (written examination + bonus points)
Grades Chalmers | Points | Grades GU | Points |
- | < 15 | U | < 15 |
3 | 15-20 | G | 15-27 |
4 | 21-27 | VG | > 27 |
5 | > 27 |
Changes compared to the last occasion
1. New homework 2.
2. New comp. ex. 2,3.
3. Modified comp.ex.4.
Deadlines for computer exercises and homeworks:
Homework 1 and comp. ex. 1: 13 September
Homework 2: 20 September
Homework 3 and comp.ex. 2: 4 October
Homework 4 and comp.ex. 3: 11 October
Comp.ex. 4: 18 October
Announcement of the course "Introduction to Inverse and Ill-posed Problems", 7.5 Hp
- Lecture 1
Introduction and organization of the course. Introduction to linear algebra and numerical linear algebra. If this looks unfamiliar you should better consult you former literature in linear algebra and refresh your knowledges. We will concentrate on the three building bricks in Numerical Linear Algebra (NLA) : (1) Linear Systems, (2) Overdetermined systems by least squares, and (3) Eigenproblems. The building bricks serve as subproblems also in nonlinear computations by the idea of linearization. We considered example of application of linear systems: image compression using SVD, image deblurring. We introduced some basic concepts and notations: transpose, lower and upper triangular matrices, singular matrices, symmetric and positive definite matrix, conjugate transpose matrix, row echelon form, rank, cofactor.
- Lecture 2
IEEE system and floating-point numbers. We will discuss computer exercise 1 and perturbation theory in polynomial evaluation. - Lecture 2
- Lecture 3
We will discuss why we need norms: (1) to measure errors, (2) to measure stability, (3) for convergence criteria of iterative methods. Sherman - Morrison formula. System of linear equations. Gaussian elimination, LU-factorization. Gaussian elimination (and LU factorization) by outer products (elementary transformations). Pivoting, partial and total. - Lecture 3
- Lecture 4
We will discuss the need of pivoting and uniqueness of factorization of A. Different versions of algorithms of Gaussian elimination. Error bounds for the solution Ax=b. Roundoff error analysis in LU factorization. Estimating condition numbers. Hagers's algorithm. - Lecture 4
- Lecture 5
Componentwise error estimates. Improving the Accuracy of a Solution for Ax=b: Newton's method and equilibration technique. Convergence of the Newton's method. Real Symmetric Positive Definite Matrices. Cholesky algorithm. Example: how to solve Poisson's equation on a unit square using LU factorization. - Lecture 5
- Lecture 6
Band Matrices, example: application of Cholesky decomposition in the solution of ODE. Continuation of considering example: application of Cholesky decomposition in the solution of ODE. Linear least squares problems. Introduction and applications. The normal equations. - Lecture 6
- Lecture 7
- Example: Polynomial fitting to curve. Solution of nonlinear least squares problems: examples.
QR and Singular value decomposition (SVD). QR and SVD for linear least squares problems. We discussed computer exercise 2. - Lecture 7
- Lecture 8
- Example of application of linear systems: image compression using SVD in Matlab. Least squares and classification algorithms. We discussed computer exercise 3.
- Lecture 8
- Lecture 9
- Householder transformations and Givens rotations. QR-factorization by Householder transformation and Givens rotation. Examples of performing Householder transformation for QR decomposition and tridiagonalization of matrix. Moore-Penrose pseudoinverse. Rank-Deficient Least Squares Problems. Least squares and classification algorithms.
Introduction to spectral theory, eigenvalues, right and left eigenvectors. Similar matrices. Defective eigenvalues. Canonical forms: Jordan form. - Lecture 9
- Lecture 10
Canonical forms: Jordan form and Schur form, real Schur form. Gerschgorin's theorem. Perturbation theory, Bauer-Fike theorem. Discussed algorithms for the non-symmetric eigenproblems: power method, inverse iteration, inverse iteration with shift. - Lecture 10
- Lecture 11
Discussed algorithms for the non-symmetric eigenproblems: inverse iteration with shift, orthogonal iteration, QR iteration and QR-iteration with shift. Hessenberg matrices, preservation of Hessenberg form. Hessenberg reduction. Tridiagonal and bidiagonal reduction. Regular Matrix Pencils and Weierstrass Canonical Form. - Lecture 11
- Lecture 12
Regular Matrix Pencils and Weierstrass Canonical Form. Singular Matrix Pencils and the Kronecker Canonical Form. Application of Jordan and Weierstrass Forms to Differential Equations. Symmetric eigenproblems Perturbation theory: Weyl's theorem. Corollary regarding similar result for singular values. Application of Weyl's theorem: Error bounds for eigenvalues computed by a stable method. Courant-Fischer (CF) theorem. Inertia, Sylvester's inertia theorem. Definite Pencils. Theorem that the Rayleigh quotient is a good approximation to an eigenvalue. Algorithms for the Symmetric eigenproblem: Tridiagonal QR iteration, Rayleigh quitient iteration. - Lecture 12
- Lecture 13
Theorem that the Rayleigh quotient is a good approximation to an eigenvalue. Algorithms for the Symmetric eigenproblem: Tridiagonal QR iteration, Rayleigh quitient iteration, Divide-and-conquer algorithm. QR iteration with Wilkinson's shift. Divide-and-conquer, bisection and inverse iteration, different versions of Jacobi's method. - Lecture 13
- Lecture 14
Algorithms for symmetric matrices (continuation): different versions of Jacobi's method. Algorithms for the SVD: QR iteration and Its Variations for the Bidiagonal SVD. The basic iterative methods (Jacobi, Gauss-Seidel and Successive overrelaxation (SOR)) for solution of linear systems. Jacobi, Gauss-Seidel and SOR for the solution of the Poisson's equation in two dimension. Study of Convergence of Jacobi, Gauss-Seidel and SOR. Introduction to the Krylov subspace methods. Conjugate gradient algorithm. Preconditioning for Linear Systems. Preconditioned conjugate gradient algorithm. Common preconditioners. - Lecture 14
Homeworks
- To pass this course you should do 2 compulsory home assignments before the final exam. Choose any 2 of 4 assignments presented here:
- Homeworks
- Homeworks should do done individually (not in groups).
- Sent pdf file with your assignment to my e-mail larisa@chalmers.se
before deadline (see the course page "Program" for deadlines for every home assignment). Handwritten home assignments can be left in the red box located beside my office.
Recommended exercises in the course book
Chapters | Exercises |
---|---|
8 | 8.7, 8.10, 8.16 |
9 | 9.5, 9.7, 9.8, 9.9, 9.11, 9.12, 9.14, 9.15 |
11 | 11.1 - 11-13 |
Computer labs
- To pass this course you should do any 2 computer assignments described here
- Computer labs
- Paper with real values of parameters for comp.ex.2 (see Table II, page 345 )
- Database for grey seals (xlsx file) for comp.ex.3.
- PETSc code for solution Helmholtz equation (see Programs for download)
- Every computer exercise should be presented in the form of written report and programs written in Matlab or C++/PETSC. You can download latex-template for the report here:
- latex-template (pdf-file)
- latex-template (tex-file)
- You can work in the groups by 2 persons.
- Sent final report for every computer assignment with description of your work together with Matlab or C++/PETSc programs for testing to my e-mail before the deadline.
- Report should contain: description of used techniques, tables and figures confirming your investigations, analysis of obtained results. Summarized results should be described in the final section ``Conclusions''.
- Matlab and C++/PETSc programs for examples in the course book are available for free download: go to the link of the course book and click to ``GitHub Page with MATLAB® Source Codes'' on the bottom of this page.
Reference literature:
- Learning MATLAB, Tobin A. Driscoll. Provides a brief introduction to Matlab to the one who already knows computer programming. Available as e-book from Chalmers library.
-
Physical Modeling in MATLAB 3/E, Allen B. Downey
The book is free to download from the web. The book gives an introduction for those who have not programmed before. It covers basic MATLAB programming with a focus on modeling and simulation of physical systems.
Course summary:
Date | Details | Due |
---|---|---|