Single-Course Engelsk 5 ECTS

Algebraic Graph Theory

Overall Course Objectives

The course will provide students with the mathematical tools for applying
algebraic techniques to graph theoretic questions. The main focus will be applying linear
algebraic tools to matrices associated to graphs in order to reveal graph structure and
information about graph parameters.

See course description in Danish

Learning Objectives

  • Understand eigenvalues/eigenvectors of graphs and their role in spectral graph theory, e.g., proving bounds on graph parameters
  • Apply the spectral decomposition theorem to prove graph theoretic results
  • Apply semidefinite programming techniques to prove results about the Lovasz theta function, e.g., that it bounds the Shannon capacity
  • Prove standard results about coherent algebras and association schemes
  • Prove and apply the Erdos-Ko-Rado theorem
  • Use linear algebra and group theory to prove results about graph homomorphisms
  • Apply the Weisfeiler-Leman algorithm
  • Understand the characterization of graphs with minimum eigenvalue at least -2
  • Prove results in fractional graph theory, e.g., about fractional colorings and fractional isomorphisms

Course Content

1. Eigenvalues and eigenvectors of graphs
2. Graph homomorphisms
3. Applications of linear and semidefinite programming to graph theory
4. Erdos-Ko-Rado Theorem
5. Graph automorphisms
6. Fractional graph theory
7. Shannon capacity and the Lovasz theta function

Possible start times

  • 36 – 49 (Tues 13-17)

Recommended prerequisites

01227, Mathematical maturity

Teaching Method

Two hours of lecture followed by two hours of group work
exercises.

Faculty

See course in the course database.

Registration

Language

Engelsk

Duration

13 weeks

Institute

Compute

Place

DTU Lyngby Campus

Course code 02983
Course type PhD
Semester start Week 36
Semester end Week 49
Days Tues 13-17
Price

10.600,00 DKK

Registration