MAGIC013: Matrix Analysis

Course details

Semester

Autumn 2009
Monday, October 5th to Friday, December 11th

Hours

Live lecture hours
20
Recorded lecture hours
0
Total advised study hours
0

Timetable

Mondays
11:05 - 11:55
Fridays
11:05 - 11:55

Description

This is offered as a core course for Applied.
 
Matrix theory is an active research field, with at least international four journals devoted to the subject. It is also an important component in many areas of applied mathematics - numerical analysis, optimisation, statistics, applied probability, image processing, ...
The solution to many problems in Science, Engineering and Mathematics lies in a "matrix fact". Dennis Bernstein

Prerequisites

No prerequisites information is available yet.

Syllabus

  1. Introduction (2 lectures)
    • Matrix products - Standard product, tensor/Kronecker product, Schur product
    • Decompostions - Schur form, Real Schur form, Jordan form, Singular Value decompositions
    • Other preliminaries - Schur complement, additive and multiplicative compounds
  2. Norms (3 lectures)
    • norms on vector spaces
    • inequalities relating norms
    • matrix norms
    • unitarily invariant norms
    • numerical radius
    • perturbation theory for linear systems
  3. Gerschgorin's Thorem, Non-negative matrices and Perron-Frobenius (4 lectures)
    • diagonal dominance and Gerschgorin's Theorem
    • spectrum of stochastic and doubly stochastic matrices
    • Sinkhorn balancing
    • Perron-Frobenius Theorem
    • Matrices realted to non-negative matrices - M-matrix, P-matrix, totally positive matrices.
  4. Spectral Theory for Hermitian matrices (2 lectures)
    • Orthogonal diagonalisation
    • Interlacing and Monotonicity of Eigenvalues
    • Weyl's and the Lidskii-Weilandt inequalities
  5. Singular values and best approximation problems (2 lectures)
    • Connection with Hermitian eigenvalue problem
    • Lidskii-Weilandt - additive and multiplicative versions
    • best rank-k approximations
    • polar factorisation, closest unitary matrix, closest rectangular matrix with orthonormal columns
  6. Positive definite matrices (3 lectures)
    • Characterisations
    • Schur Product theorem
    • Determinantal inequalitties
    • semidefinite completions
    • The Loewner theory
  7. Perturbation Theory for Eigenvalues and Eigenvectors (2 lectures)
    • primarily the non-Hermitian case
  8. Functions of matrices (2 lectures)
    • equivalance of definitions of f(A)
    • approximation of/algorithms for general functions
    • special methods for particular functions (squareroot, exponential, logarithm, trig. functions)

Lecturer

Bibliography

No bibliography has been specified for this course.

Assessment

Attention needed

Assessment information will be available nearer the time.

Files

Only consortium members have access to these files.

Please log in to view course materials.

Lectures

Please log in to view lecture recordings.