Locally identifying colorings of graphs Aline Parreau Joint work - - PowerPoint PPT Presentation

locally identifying colorings of graphs
SMART_READER_LITE
LIVE PREVIEW

Locally identifying colorings of graphs Aline Parreau Joint work - - PowerPoint PPT Presentation

Locally identifying colorings of graphs Aline Parreau Joint work with Louis Esperet, Sylvain Gravier, Micka el Montassier, Pascal Ochem Turku, March 28th, 2011 ANR IDEA 1/40 Grenoble-Turku 2/40 Outline Proper Colorings Definition


slide-1
SLIDE 1

Locally identifying colorings of graphs

Aline Parreau

Joint work with Louis Esperet, Sylvain Gravier, Micka¨ el Montassier, Pascal Ochem Turku, March 28th, 2011 ANR IDEA

1/40

slide-2
SLIDE 2

Grenoble-Turku

2/40

slide-3
SLIDE 3

Outline

Proper Colorings Definition Perfect graphs Colorings of hypergraph Locally Identifying Colorings Definition Bipartite graphs Perfect graphs are not any more perfect Planar graphs

3/40

slide-4
SLIDE 4

Proper coloring

Graph G = (V , E) Proper coloring : c : V → N s.t. uv ∈ E ⇒ c(u) = c(v)

4/40

slide-5
SLIDE 5

Proper coloring

Graph G = (V , E) Proper coloring : c : V → N s.t. uv ∈ E ⇒ c(u) = c(v) 1 2 3 4 5 6 1 2 3 4 5 1

4/40

slide-6
SLIDE 6

A well-known example

4-Colour Theorem : But also : frequency allocation, scheduling,...

5/40

slide-7
SLIDE 7

A well-known example

4-Colour Theorem : But also : frequency allocation, scheduling,...

5/40

slide-8
SLIDE 8

A well-known example

4-Colour Theorem : But also : frequency allocation, scheduling,...

5/40

slide-9
SLIDE 9

Chromatic number

Chromatic number χ(G) : minimum number of colors to have a porper coloring : 1 2 3 4 5 6 1 2 3 4 5 1 χ(G) ≤ 6

6/40

slide-10
SLIDE 10

Chromatic number

Chromatic number χ(G) : minimum number of colors to have a porper coloring : 1 2 3 4 5 6 1 2 3 4 5 1 1 2 3 4 5 5 1 2 3 4 5 1 χ(G) ≤ 5

6/40

slide-11
SLIDE 11

Chromatic number

Chromatic number χ(G) : minimum number of colors to have a porper coloring : 1 2 3 4 5 6 1 2 3 4 5 1 1 2 3 2 3 2 3 3 1 2 3 1 χ(G) = 3

6/40

slide-12
SLIDE 12

A lower bound...

Clique number ω(G) : max k such that there are k vertices in G that are all connected to each other ω(G) = 3

7/40

slide-13
SLIDE 13

A lower bound...

Clique number ω(G) : max k such that there are k vertices in G that are all connected to each other ω(G) = 4

7/40

slide-14
SLIDE 14

A lower bound...

Clique number ω(G) : max k such that there are k vertices in G that are all connected to each other ω(G) = 4

For any graph G, χ(G) ≥ ω(G)

7/40

slide-15
SLIDE 15

... that is not always reached

χ(C5) = 3 but ω(C5) = 2

8/40

slide-16
SLIDE 16

... that is not always reached

1 2 1 2 3 χ(C5) = 3 but ω(C5) = 2

8/40

slide-17
SLIDE 17

... that is not always reached

1 2 1 2 3 χ(C5) = 3 but ω(C5) = 2 Mycielski graphs Mk such that χ(Mk) = k but ω(Mk) = 2

8/40

slide-18
SLIDE 18

An upper bound

∆(G) : maximum degree of G Upper bound with greedy algorithm : χ(G) ≤ ∆(G) + 1 Tight for : complete graphs, odd cycles Brook’s theorem (1941) : χ(G) ≤ ∆(G) if G is not a complete graph or an odd cycle

9/40

slide-19
SLIDE 19

It’s hard to guess !

k-Coloring is NP-complete for any k ≥ 3 Even for k = 3 and planar graphs with maximum degree 4 !

10/40

slide-20
SLIDE 20

Good graphs for coloring

Perfect graph (1963) : G is perfect if ω(H) = χ(H) for any induced subgraph H of G Examples :

  • trees (and bipartite graphs) :
  • interval graphs :

11/40

slide-21
SLIDE 21

Strong Perfect Graph Theorem

Remark : perfect graphs are stable by induced subgraphs Smallest non perfect graphs ? → odd cycles of size ≥ 5, complement of odd cycles of size ≥ 5 Strong Perfect Graph Theorem (Chudnovsky, Robertson, Seymour, Thomas 2002) : G is perfect if and only if it has no induced odd cycle or complement of odd cycle with more than 5 vertices

12/40

slide-22
SLIDE 22

A part of the big family of perfect graphs Perfect

Chordal Permutation Line of bipartite Cograph Trees k-trees Split Bipartite Interval

13/40

slide-23
SLIDE 23

k-trees

Construction of 2-trees :

  • Start with a triangle

14/40

slide-24
SLIDE 24

k-trees

Construction of 2-trees :

  • Start with a triangle
  • Choose an edge

14/40

slide-25
SLIDE 25

k-trees

Construction of 2-trees :

  • Start with a triangle
  • Choose an edge
  • Add a vertex connected to

the 2 vertices of the edge

14/40

slide-26
SLIDE 26

k-trees

Construction of 2-trees :

  • Start with a triangle
  • Choose an edge
  • Add a vertex connected to

the 2 vertices of the edge

  • Repeat the operation

14/40

slide-27
SLIDE 27

k-trees

Construction of 2-trees :

  • Start with a triangle
  • Choose an edge
  • Add a vertex connected to

the 2 vertices of the edge

  • Repeat the operation

Remark : any edge is in a triangle

14/40

slide-28
SLIDE 28

k-trees

Construction of 2-trees :

  • Start with a triangle
  • Choose an edge
  • Add a vertex connected to

the 2 vertices of the edge

  • Repeat the operation

Remark : any edge is in a triangle

14/40

slide-29
SLIDE 29

k-trees

Construction of 2-trees :

  • Start with a triangle
  • Choose an edge
  • Add a vertex connected to

the 2 vertices of the edge

  • Repeat the operation

Remark : any edge is in a triangle

14/40

slide-30
SLIDE 30

k-trees

Construction of 2-trees :

  • Start with a triangle
  • Choose an edge
  • Add a vertex connected to

the 2 vertices of the edge

  • Repeat the operation

Remark : any edge is in a triangle

14/40

slide-31
SLIDE 31

Hypergraph

Hypergraph H = (V , E) where an edge e ∈ E is a subset of vertices

E D C B A e1 = {A, E} e2 = {A, B} e3 = {B, C, D} e4 = {C, D, E}

Hypergraph

15/40

slide-32
SLIDE 32

Hypergraph

Hypergraph H = (V , E) where an edge e ∈ E is a subset of vertices

E D C B A e1 = {A, E} e2 = {A, B} e3 = {B, C, D} e4 = {C, D, E}

Hypergraph E D C B A

e4 e3 e2 e1

Corresponding bipartite graph

15/40

slide-33
SLIDE 33

Colorings of Hypergraph

c : V → N is a proper coloring of H if no edge is unicolor.

E D C B A

A 2-proper coloring 2-Hypergraph-Coloring is NP-complete

16/40

slide-34
SLIDE 34

Locally identifying coloring (lid-coloring)

In proper coloring : Two adjacent vertices have distinct colors . 1 1 2 1 2 2

Bt(u) = {v | d(u, v) ≤ t}

For any edge uv, c(B0(u)) = c(B0(v))

17/40

slide-35
SLIDE 35

Locally identifying coloring (lid-coloring)

In proper coloring : Two adjacent vertices have distinct colors . 1 1 2 1 2 2

Bt(u) = {v | d(u, v) ≤ t}

For any edge uv, c(B0(u)) = c(B0(v))

17/40

slide-36
SLIDE 36

Locally identifying coloring (lid-coloring)

In locally identifying coloring : Two adjacent vertices have distinct colors in their neighborhood. 1 1 2 1 2 2

Bt(u) = {v | d(u, v) ≤ t}

For any edge uv, c(B0(u)) = c(B0(v)) and c(B1(u)) = c(B1(v))

17/40

slide-37
SLIDE 37

Locally identifying coloring (lid-coloring)

In locally identifying coloring : Two adjacent vertices have distinct colors in their neighborhood. 1 1 2 3 4 4

1, 2, 3 2, 3, 4

Bt(u) = {v | d(u, v) ≤ t}

For any edge uv, c(B0(u)) = c(B0(v)) and c(B1(u)) = c(B1(v))

17/40

slide-38
SLIDE 38

Locally identifying coloring (lid-coloring)

In locally identifying coloring : Two adjacent vertices have distinct colors in their neighborhood. 1 1 2 3 4 4

1, 2, 3 2, 3, 4

Bt(u) = {v | d(u, v) ≤ t}

For any edge uv, c(B0(u)) = c(B0(v)) and c(B1(u)) = c(B1(v)) whenever B1(u) = B1(v)

17/40

slide-39
SLIDE 39

Locally identifying coloring (lid-coloring)

In locally identifying coloring : Two adjacent vertices have distinct colors in their neighborhood. 1 1 2 3 4 4

1, 2, 3 2, 3, 4

Bt(u) = {v | d(u, v) ≤ t}

For any edge uv, c(B0(u)) = c(B0(v)) and c(B1(u)) = c(B1(v)) whenever B1(u) = B1(v) χlid(G) : lid-chromatic number

17/40

slide-40
SLIDE 40

An example : the path

With 4 colors :

18/40

slide-41
SLIDE 41

An example : the path

With 4 colors : 1 2 3 4 1 2 3 4

18/40

slide-42
SLIDE 42

An example : the path

With 4 colors : 1 2 3 4 1 2 3 4

1, 2 1, 2, 3 2, 3, 4 1, 3, 4 1, 2, 4 1, 2, 3 2, 3, 4 3, 4

So : χlid(Pk) ≤ 4

18/40

slide-43
SLIDE 43

An example : the path

With 4 colors : 1 2 3 4 1 2 3 4

1, 2 1, 2, 3 2, 3, 4 1, 3, 4 1, 2, 4 1, 2, 3 2, 3, 4 3, 4

So : χlid(Pk) ≤ 4 Is it possible with 3 colors ?

18/40

slide-44
SLIDE 44

An example : the path

With 4 colors : 1 2 3 4 1 2 3 4

1, 2 1, 2, 3 2, 3, 4 1, 3, 4 1, 2, 4 1, 2, 3 2, 3, 4 3, 4

So : χlid(Pk) ≤ 4 Is it possible with 3 colors ?

18/40

slide-45
SLIDE 45

An example : the path

With 4 colors : 1 2 3 4 1 2 3 4

1, 2 1, 2, 3 2, 3, 4 1, 3, 4 1, 2, 4 1, 2, 3 2, 3, 4 3, 4

So : χlid(Pk) ≤ 4 Is it possible with 3 colors ? 1 2

1, 2

18/40

slide-46
SLIDE 46

An example : the path

With 4 colors : 1 2 3 4 1 2 3 4

1, 2 1, 2, 3 2, 3, 4 1, 3, 4 1, 2, 4 1, 2, 3 2, 3, 4 3, 4

So : χlid(Pk) ≤ 4 Is it possible with 3 colors ? 1 2

1, 2

3

1, 2, 3

18/40

slide-47
SLIDE 47

An example : the path

With 4 colors : 1 2 3 4 1 2 3 4

1, 2 1, 2, 3 2, 3, 4 1, 3, 4 1, 2, 4 1, 2, 3 2, 3, 4 3, 4

So : χlid(Pk) ≤ 4 Is it possible with 3 colors ? 1 2

1, 2

3

1, 2, 3

2

2, 3

18/40

slide-48
SLIDE 48

An example : the path

With 4 colors : 1 2 3 4 1 2 3 4

1, 2 1, 2, 3 2, 3, 4 1, 3, 4 1, 2, 4 1, 2, 3 2, 3, 4 3, 4

So : χlid(Pk) ≤ 4 Is it possible with 3 colors ? 1 2

1, 2

3

1, 2, 3

2

2, 3

1

1, 2, 3

18/40

slide-49
SLIDE 49

An example : the path

With 4 colors : 1 2 3 4 1 2 3 4

1, 2 1, 2, 3 2, 3, 4 1, 3, 4 1, 2, 4 1, 2, 3 2, 3, 4 3, 4

So : χlid(Pk) ≤ 4 Is it possible with 3 colors ? 1 2

1, 2

3

1, 2, 3

2

2, 3

1

1, 2, 3

2 3 2

1, 2 1, 2, 3 2, 3 2, 3

χlid(Pk) = 3 ⇔ k is odd

18/40

slide-50
SLIDE 50

Related works

With edge colorings :

  • Vertex-distinguishing edge colorings (Observability of a graph)

(Hornak et al, 95’),

  • Adjacent vertex-distinguishing edge colorings (Zhang et al,

02’) With total colorings :

  • Adjacent vertex-distinguishing total colorings (Zhang, 05’)

19/40

slide-51
SLIDE 51

Link with chromatic number

χlid(G) ≥ χ(G) Do we need much more than χ(G) colors ?

20/40

slide-52
SLIDE 52

Link with chromatic number

χlid(G) ≥ χ(G) Do we need much more than χ(G) colors ?

20/40

slide-53
SLIDE 53

Link with chromatic number

χlid(G) ≥ χ(G) Do we need much more than χ(G) colors ? An example with χ(G) = 3 and χlid(G) ≥ k

20/40

slide-54
SLIDE 54

Link with chromatic number

χlid(G) ≥ χ(G) Do we need much more than χ(G) colors ? 1 1 An example with χ(G) = 3 and χlid(G) ≥ k

20/40

slide-55
SLIDE 55

Link with chromatic number

χlid(G) ≥ χ(G) Do we need much more than χ(G) colors ? 1 1 2 3 An example with χ(G) = 3 and χlid(G) ≥ k

20/40

slide-56
SLIDE 56

Link with chromatic number

χlid(G) ≥ χ(G) Do we need much more than χ(G) colors ? 1 1 2 3 1, 2, 3 ← 1, 2, 3 ← An example with χ(G) = 3 and χlid(G) ≥ k

20/40

slide-57
SLIDE 57

Link with chromatic number

χlid(G) ≥ χ(G) Do we need much more than χ(G) colors ? An example with χ(G) = 3 and χlid(G) ≥ k χlid is not bounded by a function of χ But...

20/40

slide-58
SLIDE 58

Link with maximum degree

We have : χlid(G) ≤ χ(G 3) This implies : χlid(G) ≤ ∆(G)3 − ∆(G)2 + ∆(G) + 1

21/40

slide-59
SLIDE 59

Link with maximum degree

We have : χlid(G) ≤ χ(G 3) This implies : χlid(G) ≤ ∆(G)3 − ∆(G)2 + ∆(G) + 1 We know only graph that needs ∆(G)2 + ∆(G) + 1

21/40

slide-60
SLIDE 60

Transition Slide

No bounds with χ for general graphs..

What about “good classes” for proper colorings ?

22/40

slide-61
SLIDE 61

Perfect graphs Perfect

Chordal Permutation Line of bipartite Cograph Trees k-trees Split Bipartite Interval Trees Bipartite ? ?

23/40

slide-62
SLIDE 62

An amazing fact about bipartite graphs

G connected graph :

  • χlid(G) = 1 ⇒ G is a single vertex

24/40

slide-63
SLIDE 63

An amazing fact about bipartite graphs

G connected graph :

  • χlid(G) = 1 ⇒ G is a single vertex
  • χlid(G) = 2 ⇒ G is just an edge

1 1, 2 2 1, 2

24/40

slide-64
SLIDE 64

An amazing fact about bipartite graphs

G connected graph :

  • χlid(G) = 1 ⇒ G is a single vertex
  • χlid(G) = 2 ⇒ G is just an edge

1 1, 2 3 2 1, 2

24/40

slide-65
SLIDE 65

An amazing fact about bipartite graphs

G connected graph :

  • χlid(G) = 1 ⇒ G is a single vertex
  • χlid(G) = 2 ⇒ G is just an edge

1 1, 2 3 2 1, 2

  • χlid(G) = 3 ⇒ G is a triangle or a bipartite graph :

→ Partition vertices with the number of colors they see

24/40

slide-66
SLIDE 66

Bipartite graphs

L0 L1 L2 L3 L4

25/40

slide-67
SLIDE 67

Bipartite graphs

L0 L1 L2 L3 L4 1 2 3 4 1 → → → → →

1, 2 1, 2, 3 2, 3, 4 or 2, 3 1, 3, 4 or 3, 4 1, 4 25/40

slide-68
SLIDE 68

Bipartite graphs

General bounds : 3 ≤ χlid(B) ≤ 4

26/40

slide-69
SLIDE 69

Bipartite graphs

General bounds : 3 ≤ χlid(B) ≤ 4

χlid(B) = 3 :

26/40

slide-70
SLIDE 70

Bipartite graphs

General bounds : 3 ≤ χlid(B) ≤ 4

χlid(B) = 3 : χlid(B) = 4 :

26/40

slide-71
SLIDE 71

Bipartite graphs

General bounds : 3 ≤ χlid(B) ≤ 4

χlid(B) = 3 : χlid(B) = 4 :

← ? → In general... 3-Lid-Coloring is NP-complete in bipartite graphs

26/40

slide-72
SLIDE 72

Link with 2-coloring of hypergraph

Try to color a graph with 3 colors

27/40

slide-73
SLIDE 73

Link with 2-coloring of hypergraph

Try to color a graph with 3 colors

1, 2, 3

1 2 3

27/40

slide-74
SLIDE 74

Link with 2-coloring of hypergraph

Try to color a graph with 3 colors

1, 2, 3 1, 2 1, 3

1 1 1 2 3

27/40

slide-75
SLIDE 75

Link with 2-coloring of hypergraph

Try to color a graph with 3 colors

1, 2, 3 1, 2 1, 3 1, 2 1, 2 1, 3 1, 2, 3 1, 2, 3 1, 2, 3

2 2 3 1 1 1 1 2 3

27/40

slide-76
SLIDE 76

Link with 2-coloring of hypergraph

Try to color a graph with 3 colors

1, 2, 3 1, 2 1, 3 1, 2 1, 2 1, 3 1, 2, 3 1, 2, 3 1, 2, 3

E D C B A

E D C B A

27/40

slide-77
SLIDE 77

Link with 2-coloring of hypergraph

Try to color a graph with 3 colors

1, 2, 3 1, 2 1, 3 1, 2 1, 2 1, 3 1, 2, 3 1, 2, 3 1, 2, 3

E D C B A

E D C B A

  • 3-Lid-Coloring in bipartite graph is NP-Complete
  • Polynomial if B regular, if B is planar with maximum degree

3, if B is a tree.

27/40

slide-78
SLIDE 78

Perfect graphs Perfect

Chordal Permutation Line of bipartite Cograph Trees k-trees Split Bipartite Interval Trees Bipartite k-trees

≤ 4 = 2ω ≤ 4 = 2ω

?

28/40

slide-79
SLIDE 79

To perfect graph : k-trees

Lid-coloring of 2-trees with 6 colors :

1 2 3

  • Color the triangle with

colors 1, 2, 3

  • Step :

i

29/40

slide-80
SLIDE 80

To perfect graph : k-trees

Lid-coloring of 2-trees with 6 colors :

1 2 3 5

  • Color the triangle with

colors 1, 2, 3

  • Step :

i i + 3[6]

  • We always have :

◮ proper coloring ◮ no edge (i, i + 3) 29/40

slide-81
SLIDE 81

To perfect graph : k-trees

Lid-coloring of 2-trees with 6 colors :

1 2 3 5 6

  • Color the triangle with

colors 1, 2, 3

  • Step :

i i + 3[6]

  • We always have :

◮ proper coloring ◮ no edge (i, i + 3) 29/40

slide-82
SLIDE 82

To perfect graph : k-trees

Lid-coloring of 2-trees with 6 colors :

1 2 3 5 6 4

  • Color the triangle with

colors 1, 2, 3

  • Step :

i i + 3[6]

  • We always have :

◮ proper coloring ◮ no edge (i, i + 3) 29/40

slide-83
SLIDE 83

To perfect graph : k-trees

Lid-coloring of 2-trees with 6 colors :

1 2 3 5 6 4 4

  • Color the triangle with

colors 1, 2, 3

  • Step :

i i + 3[6]

  • We always have :

◮ proper coloring ◮ no edge (i, i + 3) 29/40

slide-84
SLIDE 84

To perfect graph : k-trees

Lid-coloring of 2-trees with 6 colors :

1 2 3 5 6 4 4 6 5 3 4 2 3

  • Color the triangle with

colors 1, 2, 3

  • Step :

i i + 3[6]

  • We always have :

◮ proper coloring ◮ no edge (i, i + 3) 29/40

slide-85
SLIDE 85

To perfect graph : k-trees

We can extend the construction to k-trees : → A k-tree has lid-chromatic number at most 2k + 2 This bound is sharp : Pk

2k+2

1 2 3 4 5 6 7 8

30/40

slide-86
SLIDE 86

To perfect graph : k-trees

We can extend the construction to k-trees : → A k-tree has lid-chromatic number at most 2k + 2 This bound is sharp : Pk

2k+2

1 2 3 4 5 6 7 8 1 2 3 4 Complete graph

30/40

slide-87
SLIDE 87

To perfect graph : k-trees

We can extend the construction to k-trees : → A k-tree has lid-chromatic number at most 2k + 2 This bound is sharp : Pk

2k+2

1 2 3 4 5 6 7 8 1 2 3 4 5 Separated by vertex 5

30/40

slide-88
SLIDE 88

To perfect graph : k-trees

We can extend the construction to k-trees : → A k-tree has lid-chromatic number at most 2k + 2 This bound is sharp : Pk

2k+2

1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8

30/40

slide-89
SLIDE 89

Perfect Graphs Perfect

Chordal Permutation Line of bipartite Cograph Trees k-trees Split Bipartite Interval Trees

≤ 4 = 2ω

Bipartite

≤ 4 = 2ω

k-trees

≤ 2k + 2 = 2ω

31/40

slide-90
SLIDE 90

Perfect Graphs Perfect

Chordal Permutation Line of bipartite Cograph Trees k-trees Split Bipartite Interval Trees

≤ 4 = 2ω

Bipartite

≤ 4 = 2ω

k-trees

≤ 2k + 2 = 2ω

Split Interval Cograph

≤ 2ω − 1 ≤ 2ω − 1 ≤ 2ω

31/40

slide-91
SLIDE 91

Perfect Graphs Perfect

Chordal Permutation Line of bipartite Cograph Trees k-trees Split Bipartite Interval Trees

≤ 4 = 2ω

Bipartite

≤ 4 = 2ω

k-trees

≤ 2k + 2 = 2ω

Split Interval Cograph

≤ 2ω − 1 ≤ 2ω − 1 ≤ 2ω

Perfect

Chordal ?

?

31/40

slide-92
SLIDE 92

Perfect graphs are not any more perfect...

Question : Can we color any perfect graph G with 2ω(G) colors ?

32/40

slide-93
SLIDE 93

Perfect graphs are not any more perfect...

Question : Can we color any perfect graph G with 2ω(G) colors ? No ! V1 V2 V3

M Kk,k \ M Kk,k

V1, V2, V3 stable sets of size k

χlid ≥ k + 2 but ω = 3

32/40

slide-94
SLIDE 94

Perfect graphs are not any more perfect...

Question : Can we color any perfect graph G with 2ω(G) colors ? No ! V1 V2 V3

M Kk,k \ M Kk,k

V1, V2, V3 stable sets of size k

χlid ≥ k + 2 but ω = 3 Conjecture : We can color any chordal graph G with 2ω(G) colors

32/40

slide-95
SLIDE 95

Perfect Graphs Perfect

Chordal Permutation Line of bipartite Cograph Trees k-trees Split Bipartite Interval Trees

≤ 2ω

Bipartite ≤ 2ω k-trees

≤ 2ω

Split Interval Cograph

≤ 2ω ≤ 2ω ≤ 2ω

Perfect

Chordal NO ?

33/40

slide-96
SLIDE 96

To support the conjecture : Split graphs

Chordal graph : constructed like k-trees but the size of the clique can change

34/40

slide-97
SLIDE 97

To support the conjecture : Split graphs

Chordal graph : constructed like k-trees but the size of the clique can change Split graphs :

Kk Independant set

34/40

slide-98
SLIDE 98

To support the conjecture : Split graphs

Chordal graph : constructed like k-trees but the size of the clique can change Split graphs :

  • Bondy’s theorem : k − 1

vertices of the stable set are enough to separate the clique vertices k − 1

34/40

slide-99
SLIDE 99

To support the conjecture : Split graphs

Chordal graph : constructed like k-trees but the size of the clique can change Split graphs :

  • Bondy’s theorem : k − 1

vertices of the stable set are enough to separate the clique vertices → We can color with 2k colors → Possible with 2k − 1 colors → It’s sharp

k couleurs

k − 1 1

34/40

slide-100
SLIDE 100

Planar graphs

Is lid-chromatic number bounded for planar graphs ?

  • Worse example : 8 colors,

35/40

slide-101
SLIDE 101

Planar graphs

Is lid-chromatic number bounded for planar graphs ?

  • Worse example : 8 colors,
  • With large girth (36) bounded by 5

35/40

slide-102
SLIDE 102

Planar graphs

Is lid-chromatic number bounded for planar graphs ?

  • Worse example : 8 colors,
  • With large girth (36) bounded by 5

35/40

slide-103
SLIDE 103

Planar graphs

Is lid-chromatic number bounded for planar graphs ?

  • Worse example : 8 colors,
  • With large girth (36) bounded by 5

Outerplanar graphs :

  • General bound : 20 colors,
  • Max outerplanar graphs : ≤ 6 colors,
  • Without triangles : ≤ 8 colors,
  • Examples with at most 6 colors

35/40

slide-104
SLIDE 104

A bound for outerplanar graphs

1 2 3 4 1 L1 L2 L3 L4 L5

  • a layer = union
  • f paths,
  • 5 colors in a

layer,

  • 4 × 5 = 20

36/40

slide-105
SLIDE 105

Bound for planar graphs ?

Really large bound by Gonzcales and Pinlou (2010) More general result : Any family of graph closed by minor has lid-chromatic bounded

37/40

slide-106
SLIDE 106

A remark

  • For some subclasses of perfect graphs :

χlid(G) ≤ 2ω(G) = 2χ(G)

  • For planar graphs, worse example : χlid(G) ≤ 8 = 2χ(G)
  • For outerplanar graphs, worse example : χlid(G) ≤ 6 = 2χ(G)
  • ...

38/40

slide-107
SLIDE 107

A remark

  • For some subclasses of perfect graphs :

χlid(G) ≤ 2ω(G) = 2χ(G)

  • For planar graphs, worse example : χlid(G) ≤ 8 = 2χ(G)
  • For outerplanar graphs, worse example : χlid(G) ≤ 6 = 2χ(G)
  • ...

For which graphs do we have χlid(G) ≤ 2χ(G) ? Is it true for planar graphs ?

38/40

slide-108
SLIDE 108

Some open problems

  • Find a good bound for χlid in planar graphs
  • Prove (or disprove) conjecture for chordal graphs
  • For which graphs χlid = χ ?
  • Better bound with maximum degree ∆ ?
  • What about a global version ?

39/40

slide-109
SLIDE 109

Kiitos !

40/40