National University of Computer & Emerging Sciences
MT-3001 Graph Theory
Instructor: Miss Urooj
AUGMENTING PATHS & VERTEX COVERS
We need a way to determine if a matching is in fact maximum. We do this through the use of
alternating and augmenting paths.
OR
Example:
1
Berge’s Theorem:
We will need the following set theory definition and result about two different matchings of a
graph in our proof of Berge’s Theorem.
2
Symmetric difference:
3
4
Remark:
We can use the symmetric difference of two matchings to gain an understanding of how
they relate to each other.
In particular, we will use this concept to gain an understanding about their relative sizes,
allowing us to prove Berge’s Theorem.
Now we understand how to determine if a matching is maximum (search for augmenting
paths), we need a procedure or algorithm to actually find one.
The algorithm described below is closely related to the Hungarian Algorithm proposed by
Harold Kuhn in 1955.
Also, even though Berge’s Theorem holds for graphs that are not bipartite, the algorithm
below requires the input of a bipartite graph.
5
Remarks:
The arbitrary matching in Step (1) could be the empty matching (no edges are initially
included), though in practice starting with a quick simple matching allows for fewer
iterations of the algorithm.
You should not spend much time trying to create an initial maximum matching, but rather
choose obvious edges to include.
6
7
8
9
