MA-311 Numerical Linear Algebra
- ECTS Credits:
- 10
- Responsible department:
- Faculty of Engineering and Science
- Lecture Semester:
- Autumn
- Teaching language:
- Norwegian or English
- Duration:
- 1 term
The course is connected to the following study programs
- Industrial Mathematics, Bachelor's Programme
- Mathematics, Master's Programme
Teaching language
Norwegian or EnglishRecommended prerequisites
60 sp of Mathematics (including MA-185) and at least one subject at 200 level. Basic knowledge of a programming language (Python, MatLab, R etc.).
Course contents
Diagonalization, LU-, QR- and singular value factorization. Matrix norms. Least squares method. Iterative methods and fixed-points for solving linear equations (Jacobi and Gauss-Seidel). Preconditioning. Gradient method and Krylov-spaces. Application in e.g. Fourier-transform, partial differential equations, splines and/or machine learning.
Learning outcomes
Upon successful completion of the course, the student will be able to
-
Have knowledge about factorization techniques for matrices
-
Know how to solve linear systems with large quadratic matrices
-
Have knowledge about the least squares method for large rectangular matrices
-
Know how to find eigenvalues and eigenvectors of large matrices
-
Construct algorithms in a programming language
-
Analyse stability and convergence of computer algorithms
Examination requirements
Mandatory hand-ins must be approved. See Canvas for details.
Teaching methods
Lectures, group work, and mandatory hand-ins. Estimated workload of the course is 267 hours.
Evaluation
The person responsible for the course decides, in cooperation with student representative, the form of student evaluation and whether the course is to have a midway or end of course evaluation in accordance with the quality system for education, chapter 4.1.
Admission for external candidates
No
Offered as Single Standing Module
Yes. Subject to availability or capacity.
Assessment methods and criteria
Written, supervised, and graded exam, 5 hours.