Share to: share facebook share twitter share wa share telegram print page

Henson graph

In graph theory, the Henson graph Gi is an undirected infinite graph, the unique countable homogeneous graph that does not contain an i-vertex clique but that does contain all Ki-free finite graphs as induced subgraphs. For instance, G3 is a triangle-free graph that contains all finite triangle-free graphs.

These graphs are named after C. Ward Henson, who published a construction for them (for all i ≥ 3) in 1971.[1] The first of these graphs, G3, is also called the homogeneous triangle-free graph or the universal triangle-free graph.

Construction

To construct these graphs, Henson orders the vertices of the Rado graph into a sequence with the property that, for every finite set S of vertices, there are infinitely many vertices having S as their set of earlier neighbors. (The existence of such a sequence uniquely defines the Rado graph.) He then defines Gi to be the induced subgraph of the Rado graph formed by removing the final vertex (in the sequence ordering) of every i-clique of the Rado graph.[1]

With this construction, each graph Gi is an induced subgraph of Gi + 1, and the union of this chain of induced subgraphs is the Rado graph itself. Because each graph Gi omits at least one vertex from each i-clique of the Rado graph, there can be no i-clique in Gi.

Universality

Any finite or countable i-clique-free graph H can be found as an induced subgraph of Gi by building it one vertex at a time, at each step adding a vertex whose earlier neighbors in Gi match the set of earlier neighbors of the corresponding vertex in H. That is, Gi is a universal graph for the family of i-clique-free graphs.

Because there exist i-clique-free graphs of arbitrarily large chromatic number, the Henson graphs have infinite chromatic number. More strongly, if a Henson graph Gi is partitioned into any finite number of induced subgraphs, then at least one of these subgraphs includes all i-clique-free finite graphs as induced subgraphs.[1]

Symmetry

Like the Rado graph, G3 contains a bidirectional Hamiltonian path such that any symmetry of the path is a symmetry of the whole graph. However, this is not true for Gi when i > 3: for these graphs, every automorphism of the graph has more than one orbit.[1]

References

  1. ^ a b c d Henson, C. Ward (1971), "A family of countable homogeneous graphs", Pacific Journal of Mathematics, 38: 69–83, doi:10.2140/pjm.1971.38.69, MR 0304242.
Index: pl ar de en es fr it arz nl ja pt ceb sv uk vi war zh ru af ast az bg zh-min-nan bn be ca cs cy da et el eo eu fa gl ko hi hr id he ka la lv lt hu mk ms min no nn ce uz kk ro simple sk sl sr sh fi ta tt th tg azb tr ur zh-yue hy my ace als am an hyw ban bjn map-bms ba be-tarask bcl bpy bar bs br cv nv eml hif fo fy ga gd gu hak ha hsb io ig ilo ia ie os is jv kn ht ku ckb ky mrj lb lij li lmo mai mg ml zh-classical mr xmf mzn cdo mn nap new ne frr oc mhr or as pa pnb ps pms nds crh qu sa sah sco sq scn si sd szl su sw tl shn te bug vec vo wa wuu yi yo diq bat-smg zu lad kbd ang smn ab roa-rup frp arc gn av ay bh bi bo bxr cbk-zam co za dag ary se pdc dv dsb myv ext fur gv gag inh ki glk gan guw xal haw rw kbp pam csb kw km kv koi kg gom ks gcr lo lbe ltg lez nia ln jbo lg mt mi tw mwl mdf mnw nqo fj nah na nds-nl nrm nov om pi pag pap pfl pcd krc kaa ksh rm rue sm sat sc trv stq nso sn cu so srn kab roa-tara tet tpi to chr tum tk tyv udm ug vep fiu-vro vls wo xh zea ty ak bm ch ny ee ff got iu ik kl mad cr pih ami pwn pnt dz rmy rn sg st tn ss ti din chy ts kcg ve 
Prefix: a b c d e f g h i j k l m n o p q r s t u v w x y z 0 1 2 3 4 5 6 7 8 9 
Kembali kehalaman sebelumnya