EIDMA MINICOURSE ON STRUCTURAL GRAPH THEORY

by Robin Thomas

I will be teaching a minicourse on Structural Graph Theory at the Euler Institute for Discrete Mathematics in Eindhoven, the Nethelands during the week June 6-10, 2005.

General prerequisites: Basic notions and theorems of graph theory such as the max-flow min-cut theorem, Menger's theorem, Hall's theorem (matchings in bipartite graphs), Tutte's 1-factor theorem, Brooks' theorem, Vizing's theorem, Planar graphs, Euler's formula, Kuratowski's theorem. The standard references are [B, D].

Recommended reading: Participants may wish to familiarize themselves with specific concepts and notions ahead of each lecture. Please follow the links below for details. The book [D] covers many of those; other useful sources are [LP, RR, MT]. Those interested are welcome to look at transparencies for some of my earlier talks. Of these, the NSF-CBMS lecture series is the most relevant. I also have various lecture notes: on tree-decompositions, on perfect graphs, and on graph structure theory.

Exercises: Here is a set of introductory exercises. Please follow the links below for exercises associated with each lecture.

Lecture schedule:
  1. Planar graphs and graphs on surfaces. Review of basic properties of planar graphs, Kuratowski's theorem and the uniqueness of planar embeddings. Brief discussion of Steinitz's theorem, circle packing, Schnyder's theorem and drawing on a grid, and geometric graphs. Wye-delta transformations and applications. Planar separators. The emergence of planarity in graph structure theory, the two-paths problem. Graphs on surfaces, representativity (face-width). More information...
  2. Tree-decompositions of graphs I. Path-width, tree-width and applications to structure theory and the design of algorithms. The concept of a haven (bramble) as an obstacle to low tree-width and its use. Excluding a planar graph. The Erdos-Posa property. See also my notes on tree-decompositions. More information...
  3. Tree-decompositions of graphs II. Branch-width and algorithms. The graph minor theorem and its applications. More information...
  4. Excluded minor theorems and well-quasi-ordering. Seymour's splitter theorem and its applications. Analogues for higher connectivity and their applications. Separators. Typical subgraphs of large graphs. Excluded minor theorems for infinite graphs (time permitting). Well-quasi-ordering finite trees and extension to graphs of bounded tree-width. The finitness of obstruction sets for surface embeddings. More information...
  5. The Four-Color Theorem. History, equivalent formulations and an outline of a proof. What are the prospects for finding a computer-free proof? More information... 
  6. Beyond the Four-Color Theorem. Unique coloring, edge-coloring, packing T-joins, nowhere-zero flows, cycle double covers, Hadwiger's conjecture. Jorgensen's conjecture. Extremal problems for minors. Linkages. More information... 
  7. Matching structure, Pfaffian orientations and digraph structure I. Edmonds' matching theorem, the linear hull of perfect matchings, the matching structure: decomposition into bricks and braces, the matching lattice, Pfaffian orientations and their use, Pfaffian orientations of bipartite graphs, Polya's permanent problem, the even directed cycle problem, sign-nonsingular matrices, applications in economics. More information... 
  8. Matching structure, Pfaffian orientations and digraph structure II. Directed tree-width and digraph structure. Generating bricks, Pfaffian orientations of bricks, relation to parity of crossing numbers and signs of edge-colorings. Matching width and its relation to directed tree-width. More information coming.
  9. Perfect graphs. Examples of perfect graphs, communication complexity, graph entropy and applications, the perfect graph theorem of Lovasz, polyhedral aspects, relation to geometric optimization, graph entropy and application to sorting, the imperfection ratio and application to the channel assignment problem, stable matchings and the theorem of Gale and Shapley, 2-joins, skew partitions and the Strong Perfect Graph Theorem. More information.... 
  10. To be determined. The possibilities include: More on any of the above topics, spatial embeddings of graphs, planarity in linear time, the structure of infinite graphs.
Relevant books:
[B] B. Bollobas, Modern Graph Theory
[D] R. Diestel, Graph Theory, 2nd edition
[LP] L. Lovasz and M. Plummer, Matching theory
[MT] B. Mohar and C. Thomassen, Graphs and surfaces
[RR] A. Ramirez Alfonsin and B. Reed, Perfect graphs