The module descriptors for this programme can be found below.

Modules shown are for the current academic year and are subject to change depending on your year of entry.

Please note that the curriculum of this programme is currently being reviewed as part of a College-wide process to introduce a standardised modular structure. As a result, the content and assessment structures of this course may change for your year of entry. We therefore recommend that you check this course page before finalising your application and after submitting it as we will aim to update this page as soon as any changes are ratified by the College.

Find out more about the limited circumstances in which we may need to make changes to or in relation to our courses, the type of changes we may make and how we will tell you about changes we have made.

Computational Linear Algebra

Module aims

The module provides the student with both a theoretical and practical understanding of the standard algorithms for solving simultaneous linear equations (Ax=b), as well as of the challenges involved. It also provides the student with both a theoretical and practical understanding of the standard algorithms for efficient solution of eigenvalue problem and singular value decomposition and their applications in data compression and Principal Component Analysis (PCA).
 

Learning outcomes

  • At completion of the course, students should be able to:
     
    • Use state-of-the-art iterative methods, including sensitivity analysis of eigenvalues and eigenvectors to adjust response of a system.
    • Regularise and apply SVD to ill-posed problems
    • Utilize PCA and its application in data reduction and image processing.
    • Use Chebyshev acceleration, conjugate gradient method and GMRES as optimisation algorithms.
    • Define and implement the classical iterative algorithms
    • Implement Gaussian elimination with pivoting, LU and Cholesky factorisations as a numerical algorithm
     
     
     

Module syllabus

1 Introduction to the course; review of vector spaces, vector and matrix norms; floating point numbers and arithmetic.

2 Backward stability, accuracy of approximate solutions to linear systems, condition number; introduction to direct methods for linear systems; forward & backward substitution for triangular systems.

3 Gaussian elimination and LU factorisation; pivoting; stability of LU factorization.
 
4 LU factorisation for banded matrices; Cholesky factorisation; summary of direct methods and examples.
 
5 Splitting iterative methods for linear systems: general theory and convergence criteria, classical iterative methods, examples and comparison with direct methods.
 
6 Eigenvalue problems, algorithms.
 
7 Applications of Eigenvalue analysis: Sensitivity analysis
 
8 The Singular Value Decomposition, algorithms.
 
9 Application of SVD : Regularization, Noise reduction
 
10 PCA analysis and data reduction
 
11 Chebyshev acceleration, conjugate gradient method, GMRES.
 
12 Parallel algorithms and multigrid methods
 
 

Pre-requisites

Basic linear algebra (basis vectors, eigenvectors).

Teaching methods

The content of the module is delivered mostly by lectures and practical exercise in tutorial sessions.
Learning will be reinforced through tutorial question sheets.

Assessments

This module presents opportunities for both formative and summative assessment. You will be formatively assessed through progress tests and tutorial sessions. You will have additional opportunities to self-assess your learning via tutorial problem sheets. You will be summatively assessed through a 2-hour written closed-book examination.

You will receive feedback on examinations in the form of an examination feedback report on the performance of the entire cohort. You will receive feedback on your performance whilst undertaking tutorial exercises, during which you will also receive instructions on the correct solution to tutorial problems.
Further individual feedback will be available to you on request via this module’s online feedback forum, through staff office hours and discussions with tutors.
 
Assessment type Assessment description Weighting Pass mark
Examination Closed-book exam 100% 50%