Hadwinger's guess

from Wikipedia, the free encyclopedia
The one as a minor of a graph, for the coloring of which colors are required.

In graph theory , the presumption of Hadwiger , also abbreviated Hadwiger conjecture refers to a relationship between colorability of graphs and the occurrence of complete minors ago. What you are saying is that a graph that has no valid coloring with less than colors has a -minor. In brief: . The term is often used as an abbreviation .

The conjecture was made by Hugo Hadwiger in 1943 and is an open problem in mathematics . Béla Bollobás , Paul Allen Catlin and Paul Erdős (1980) called it "one of the deepest unsolved problems in graph theory".

The conjecture is closely related to the four-color theorem , which - taking into account Kuratowski's theorem and other theorems of graph theory - is equivalent to it for , i.e. with , and at the same time provides the basis for the only known proof of Neil Robertson , Paul Seymour and Robin Thomas were able to show in 1993 that Hadwinger's conjecture for is also equivalent to the four-color theorem.

The conjecture includes the conclusion from the Robertson-Seymour theorem, proved in 2004 , that the class of graphs whose minors are all -colourable is characterized by a finite set of forbidden minors. Hadwinger's conjecture is that this set only contains the -minor.

As a tightening of Hadwiger 's conjecture , the Hajós conjecture , which does not require the as a minor , but even as a topological minor . This conjecture is correct, open and wrong for all major ones, but this has no effect on Hadwiger's conjecture.

state of things

  • Hadwiger's conjecture has so far not been proven and only partial results are available.
  • For the Hadwiger conjecture is equivalent to the four-color theorem ( Wagner's equivalence theorem ) and also for .
  • It is therefore shown for the cases .
  • From the correctness of the partial assumption follows for each natural number the accuracy of the partial assumption always . The difficulty increases as the index increases.
  • In 1980, Bela Bollobas, Paul Catlin, and Paul Erdős used probabilistic methods to show that the conjecture (in a specific sense) is correct for almost every graph .
  • In 2009 Maria Chudnovsky and Alexandra Ovetsky Fradkin were able to show that a weakened version of the Hadwiger conjecture is correct for the class of claw-free graphs, thus extending the result of Seymour, Robertsen and Thomas that the conjecture is correct for all edge graphs .
  • In 2012 Dominic van der Zypen constructed a graph H with a chromatic number ω, but without an infinite complete minor. This shows that Hadwinger's conjecture in the countable-infinite does not hold.

literature

Individual evidence

  1. a b B. Bollobás , P. Catlin and P. Erdős : Hadwiger's conjecture is true for almost every graph. (PDF) In: European Journal on Combinatorics , 1980, 1, pp. 195-199.
  2. ^ A b N. Robertson , P. Seymour , R. Thomas : Hadwiger's conjecture for -free graphs. In: Combinatorica , 1993, 13, pp. 279-361 doi: 10.1007 / BF01202354 .
  3. ^ Rudolf Halin: Graphentheorie I. 1980, p. 268 ff., 274–275
  4. ^ Reinhard Diestel: Graph Theory. 2005, pp. 172-175
  5. Rudolf Halin: Graphentheorie I. 1980, p. 268 ff., P. 275.
  6. This does not mean that the partial presumption is correct above a certain index . See discussion: Hadwinger's Conjecture !
  7. ^ Maria Chudnovsky, Alexandra Ovetsky Fradkin: An approximate version of Hadwiger's conjecture for claw-free graphs. In: Journal of Graph Theory. 2009, p. N / a, doi : 10.1002 / jgt.20425 .
  8. Dominic van der Zypen: arxiv : 1212.3093 2012.