Cop and Robber Game and Hyperbolicity J. Chalopin 1 V. Chepoi 1 . - - PowerPoint PPT Presentation

cop and robber game and hyperbolicity
SMART_READER_LITE
LIVE PREVIEW

Cop and Robber Game and Hyperbolicity J. Chalopin 1 V. Chepoi 1 . - - PowerPoint PPT Presentation

Cop and Robber Game and Hyperbolicity J. Chalopin 1 V. Chepoi 1 . Papasoglu 2 T. Pecatte 3 P 1 LIF , CNRS & Aix-Marseille Universit 2 Mathematical Institute, University of Oxford 3 NS de Lyon Sminaire GAC 23 fvrier 2017


slide-1
SLIDE 1

Cop and Robber Game and Hyperbolicity

  • J. Chalopin1
  • V. Chepoi1

P . Papasoglu2

  • T. Pecatte3

1LIF

, CNRS & Aix-Marseille Université

2Mathematical Institute, University of Oxford 3ÉNS de Lyon

Séminaire GAC – 23 février 2017

23/02/2017 Cop and Robber Game and Hyperbolicity 1/17

slide-2
SLIDE 2

Cop & Robber Game

A game between one cop C and one robber R on a graph G Initialization:

◮ C chooses a vertex ◮ R chooses a vertex

Step-by-step:

◮ C traverses at most 1 edge ◮ R traverses at most 1 edge

Winning Condition:

◮ C wins if it is on the same vertex as R ◮ R wins if it can avoid C forever

23/02/2017 Cop and Robber Game and Hyperbolicity 2/17

slide-3
SLIDE 3

Cop & Robber Game

A game between one cop C and one robber R on a graph G Initialization:

◮ C chooses a vertex ◮ R chooses a vertex

Step-by-step:

◮ C traverses at most 1 edge ◮ R traverses at most 1 edge

Winning Condition:

◮ C wins if it is on the same vertex as R ◮ R wins if it can avoid C forever

C

23/02/2017 Cop and Robber Game and Hyperbolicity 2/17

slide-4
SLIDE 4

Cop & Robber Game

A game between one cop C and one robber R on a graph G Initialization:

◮ C chooses a vertex ◮ R chooses a vertex

Step-by-step:

◮ C traverses at most 1 edge ◮ R traverses at most 1 edge

Winning Condition:

◮ C wins if it is on the same vertex as R ◮ R wins if it can avoid C forever

C R

23/02/2017 Cop and Robber Game and Hyperbolicity 2/17

slide-5
SLIDE 5

Cop & Robber Game

A game between one cop C and one robber R on a graph G Initialization:

◮ C chooses a vertex ◮ R chooses a vertex

Step-by-step:

◮ C traverses at most 1 edge ◮ R traverses at most 1 edge

Winning Condition:

◮ C wins if it is on the same vertex as R ◮ R wins if it can avoid C forever

R C

23/02/2017 Cop and Robber Game and Hyperbolicity 2/17

slide-6
SLIDE 6

Cop & Robber Game

A game between one cop C and one robber R on a graph G Initialization:

◮ C chooses a vertex ◮ R chooses a vertex

Step-by-step:

◮ C traverses at most 1 edge ◮ R traverses at most 1 edge

Winning Condition:

◮ C wins if it is on the same vertex as R ◮ R wins if it can avoid C forever

C R

23/02/2017 Cop and Robber Game and Hyperbolicity 2/17

slide-7
SLIDE 7

Cop & Robber Game

A game between one cop C and one robber R on a graph G Initialization:

◮ C chooses a vertex ◮ R chooses a vertex

Step-by-step:

◮ C traverses at most 1 edge ◮ R traverses at most 1 edge

Winning Condition:

◮ C wins if it is on the same vertex as R ◮ R wins if it can avoid C forever

R C

23/02/2017 Cop and Robber Game and Hyperbolicity 2/17

slide-8
SLIDE 8

Cop & Robber Game

A game between one cop C and one robber R on a graph G Initialization:

◮ C chooses a vertex ◮ R chooses a vertex

Step-by-step:

◮ C traverses at most 1 edge ◮ R traverses at most 1 edge

Winning Condition:

◮ C wins if it is on the same vertex as R ◮ R wins if it can avoid C forever

C R

23/02/2017 Cop and Robber Game and Hyperbolicity 2/17

slide-9
SLIDE 9

Cop & Robber Game

A game between one cop C and one robber R on a graph G Initialization:

◮ C chooses a vertex ◮ R chooses a vertex

Step-by-step:

◮ C traverses at most 1 edge ◮ R traverses at most 1 edge

Winning Condition:

◮ C wins if it is on the same vertex as R ◮ R wins if it can avoid C forever

R C

23/02/2017 Cop and Robber Game and Hyperbolicity 2/17

slide-10
SLIDE 10

Cop-win graphs are dismantlable graphs

A graph G is cop-win if C can win whatever R does

Theorem (Nowakowski and Winkler; Quilliot ’83)

A graph G is cop-win iff there exists a dismantling order v1, v2, . . . , vn such that ∀i > 1, ∃j < i, N[vi, Gi] ⊆ N[vj] Gi: graph induced by Xi = {v1, v2, . . . , vi} Examples of cop-win graphs: trees, cliques, chordal graphs, bridged graphs

v6 v3 v2 v4 v5 v1

23/02/2017 Cop and Robber Game and Hyperbolicity 3/17

slide-11
SLIDE 11

Cop-win graphs are dismantlable graphs

A graph G is cop-win if C can win whatever R does

Theorem (Nowakowski and Winkler; Quilliot ’83)

A graph G is cop-win iff there exists a dismantling order v1, v2, . . . , vn such that ∀i > 1, ∃j < i, N[vi, Gi] ⊆ N[vj] Gi: graph induced by Xi = {v1, v2, . . . , vi} Examples of cop-win graphs: trees, cliques, chordal graphs, bridged graphs

v3 v2 v4 v5 v1

23/02/2017 Cop and Robber Game and Hyperbolicity 3/17

slide-12
SLIDE 12

Cop-win graphs are dismantlable graphs

A graph G is cop-win if C can win whatever R does

Theorem (Nowakowski and Winkler; Quilliot ’83)

A graph G is cop-win iff there exists a dismantling order v1, v2, . . . , vn such that ∀i > 1, ∃j < i, N[vi, Gi] ⊆ N[vj] Gi: graph induced by Xi = {v1, v2, . . . , vi} Examples of cop-win graphs: trees, cliques, chordal graphs, bridged graphs

v3 v2 v4 v1

23/02/2017 Cop and Robber Game and Hyperbolicity 3/17

slide-13
SLIDE 13

Cop-win graphs are dismantlable graphs

A graph G is cop-win if C can win whatever R does

Theorem (Nowakowski and Winkler; Quilliot ’83)

A graph G is cop-win iff there exists a dismantling order v1, v2, . . . , vn such that ∀i > 1, ∃j < i, N[vi, Gi] ⊆ N[vj] Gi: graph induced by Xi = {v1, v2, . . . , vi} Examples of cop-win graphs: trees, cliques, chordal graphs, bridged graphs

v3 v2 v1

23/02/2017 Cop and Robber Game and Hyperbolicity 3/17

slide-14
SLIDE 14

Cop-win graphs are dismantlable graphs

A graph G is cop-win if C can win whatever R does

Theorem (Nowakowski and Winkler; Quilliot ’83)

A graph G is cop-win iff there exists a dismantling order v1, v2, . . . , vn such that ∀i > 1, ∃j < i, N[vi, Gi] ⊆ N[vj] Gi: graph induced by Xi = {v1, v2, . . . , vi} Examples of cop-win graphs: trees, cliques, chordal graphs, bridged graphs

v2 v1

23/02/2017 Cop and Robber Game and Hyperbolicity 3/17

slide-15
SLIDE 15

Cop-win graphs are dismantlable graphs

A graph G is cop-win if C can win whatever R does

Theorem (Nowakowski and Winkler; Quilliot ’83)

A graph G is cop-win iff there exists a dismantling order v1, v2, . . . , vn such that ∀i > 1, ∃j < i, N[vi, Gi] ⊆ N[vj] Gi: graph induced by Xi = {v1, v2, . . . , vi} Examples of cop-win graphs: trees, cliques, chordal graphs, bridged graphs

v1

23/02/2017 Cop and Robber Game and Hyperbolicity 3/17

slide-16
SLIDE 16

Cop & Robber Game with Speeds

A game between one cop C moving at speed s′ and one robber R moving at speed s Same game as before except that at each step

◮ C traverses at most s′ edges ◮ R traverses at most s edges

C

◮ C has speed s′ = 1 ◮ R has speed s = 2

23/02/2017 Cop and Robber Game and Hyperbolicity 4/17

slide-17
SLIDE 17

Cop & Robber Game with Speeds

A game between one cop C moving at speed s′ and one robber R moving at speed s Same game as before except that at each step

◮ C traverses at most s′ edges ◮ R traverses at most s edges

C R

◮ C has speed s′ = 1 ◮ R has speed s = 2

23/02/2017 Cop and Robber Game and Hyperbolicity 4/17

slide-18
SLIDE 18

Cop & Robber Game with Speeds

A game between one cop C moving at speed s′ and one robber R moving at speed s Same game as before except that at each step

◮ C traverses at most s′ edges ◮ R traverses at most s edges

R C

◮ C has speed s′ = 1 ◮ R has speed s = 2

23/02/2017 Cop and Robber Game and Hyperbolicity 4/17

slide-19
SLIDE 19

Cop & Robber Game with Speeds

A game between one cop C moving at speed s′ and one robber R moving at speed s Same game as before except that at each step

◮ C traverses at most s′ edges ◮ R traverses at most s edges

C R

◮ C has speed s′ = 1 ◮ R has speed s = 2

23/02/2017 Cop and Robber Game and Hyperbolicity 4/17

slide-20
SLIDE 20

Cop & Robber Game with Speeds

A game between one cop C moving at speed s′ and one robber R moving at speed s Same game as before except that at each step

◮ C traverses at most s′ edges ◮ R traverses at most s edges

C R

◮ C has speed s′ = 1 ◮ R has speed s = 2

23/02/2017 Cop and Robber Game and Hyperbolicity 4/17

slide-21
SLIDE 21

Cop & Robber Game with Speeds

A game between one cop C moving at speed s′ and one robber R moving at speed s Same game as before except that at each step

◮ C traverses at most s′ edges ◮ R traverses at most s edges

C R

◮ C has speed s′ = 1 ◮ R has speed s = 2

23/02/2017 Cop and Robber Game and Hyperbolicity 4/17

slide-22
SLIDE 22

(s, s′)-Cop-win Graphs and (s, s′)-dismantlability

A graph G is (s,s′)-cop-win if C (moving at speed s′) can win whatever R (moving at speed s) does

Remark

If s < s′, every graph is (s,s′)-cop-win

Theorem (C., Chepoi, Nisse, Vaxès ’11)

A graph G is (s,s′)-cop-win if and only if there exists a (s,s′)-dismantling order v1, v2, . . . , vn such that ∀i > 1, ∃j < i, Bs(vi, G \ vj) ∩ Xi ⊆ Bs′(vj) Xi = {v1, v2, . . . , vi}

23/02/2017 Cop and Robber Game and Hyperbolicity 5/17

slide-23
SLIDE 23

Two kinds of (s, s′)-dismantlability

An ordering v1, v2, . . . , vn of the vertices of V(G) is

◮ (s, s′)-dismantling if

∀i > 1, ∃j < i, Bs(vi, G \ vj) ∩ Xi ⊆ Bs′(vj)

◮ (s, s′)∗-dismantling if

∀i > 1, ∃j < i, Bs(vi, G) ∩ Xi ⊆ Bs′(vj)

Remarks

◮ (s, s′)-dismantling =

⇒ (s, s − 1)-dismantling if s′ < s

◮ (s, s′)∗-dismantling =

⇒ (s, s′)-dismantling

◮ (s, s − 1)-dismantling =

⇒ (s, s − 1)∗-dismantling

◮ G is (s, s)∗-dismantlable iff Gs is dismantlable

23/02/2017 Cop and Robber Game and Hyperbolicity 6/17

slide-24
SLIDE 24

δ-hyperbolic graphs

A graph (or a metric space) is δ-hyperbolic if for every four points a, b, c, d, d(a, b)+d(c, d) ≤ max{d(a, c)+d(b, d), d(a, d)+d(b, c)}+2δ The hyperbolicity δ∗ of a graph G is the minimal value of δ such that G is δ-hyperbolic

a b d c ≤ δ

23/02/2017 Cop and Robber Game and Hyperbolicity 7/17

slide-25
SLIDE 25

δ-hyperbolic graphs

A graph (or a metric space) is δ-hyperbolic if for every four points a, b, c, d, d(a, b)+d(c, d) ≤ max{d(a, c)+d(b, d), d(a, d)+d(b, c)}+2δ The hyperbolicity δ∗ of a graph G is the minimal value of δ such that G is δ-hyperbolic Examples:

◮ Trees and cliques are 0-hyperbolic ◮ Cycles are n 4-hyperbolic ◮ Square grids are √n − 1-hyperbolic ◮ Chordal graphs are 1-hyperbolic

[Brinkmann, Koolen, Moulton ’01]

a d c b

n 4 n 4 n 4 n 4 23/02/2017 Cop and Robber Game and Hyperbolicity 7/17

slide-26
SLIDE 26

δ-hyperbolic graphs

A graph (or a metric space) is δ-hyperbolic if for every four points a, b, c, d, d(a, b)+d(c, d) ≤ max{d(a, c)+d(b, d), d(a, d)+d(b, c)}+2δ The hyperbolicity δ∗ of a graph G is the minimal value of δ such that G is δ-hyperbolic Examples:

◮ Trees and cliques are 0-hyperbolic ◮ Cycles are n 4-hyperbolic ◮ Square grids are √n − 1-hyperbolic ◮ Chordal graphs are 1-hyperbolic

[Brinkmann, Koolen, Moulton ’01]

a d c b √n − 1 √n − 1

23/02/2017 Cop and Robber Game and Hyperbolicity 7/17

slide-27
SLIDE 27

δ-hyperbolic graphs

A graph (or a metric space) is δ-hyperbolic if for every four points a, b, c, d, d(a, b)+d(c, d) ≤ max{d(a, c)+d(b, d), d(a, d)+d(b, c)}+2δ The hyperbolicity δ∗ of a graph G is the minimal value of δ such that G is δ-hyperbolic

Remark

◮ The hyperbolicity of G measures how G is metrically close

from a tree

◮ There exist many definitions of δ-hyperbolicity; they are

equivalent up to a multiplicative factor

23/02/2017 Cop and Robber Game and Hyperbolicity 7/17

slide-28
SLIDE 28

Why is hyperbolicity an interesting parameter ?

A notion from Geometric Group Theory [Gromov ’87]

◮ for δ-hyperbolic group, the word problem is solvable in

linear time (it is undecidable for general groups) Some large scale graphs are known to be of small hyperbolicity

◮ the Internet topology can be embedded into a hyperbolic

space [Boguna et al. ’10]

◮ the map of the AS of the Internet has small hyperbolicity

[Cohen et al. ’ 13] Efficient algorithms exist for graphs of small hyperbolicity

◮ Greedy routing algorithms can be expected to perform very

well [Papadopoulos et al. ’09]

◮ Routing labeling schemes with small labels and small

additive error [Chepoi et al ’12]

23/02/2017 Cop and Robber Game and Hyperbolicity 8/17

slide-29
SLIDE 29

In this talk

◮ Characterization of hyperbolicity via cop and robber games

◮ δ-hyperbolic graphs are (2s, s + 2δ)-cop-win for any s ◮ (s, s − 1)-cop-win graphs are 64s2-hyperbolic

◮ An efficient algorithm to approximate the hyperbolicity of a

graph

23/02/2017 Cop and Robber Game and Hyperbolicity 9/17

slide-30
SLIDE 30

δ-hyperbolic graphs are (2s, s + 2δ)-cop-win

Proposition (from Chepoi, Estellon ’07)

Any δ-hyperbolic graph is (2s, s + 2δ)∗-dismantlable, and thus (2s, s + 2δ)-cop-win

◮ Consider any BFS ordering of V(G)

from a vertex u

◮ For all v, let v′ be a vertex on a

shortest path from v to u such that d(v, v′) = s

u v ′ v w s ≤ 2s ≤ d(u, v)

23/02/2017 Cop and Robber Game and Hyperbolicity 10/17

slide-31
SLIDE 31

δ-hyperbolic graphs are (2s, s + 2δ)-cop-win

Proposition (from Chepoi, Estellon ’07)

Any δ-hyperbolic graph is (2s, s + 2δ)∗-dismantlable, and thus (2s, s + 2δ)-cop-win Let w ∈ B2s(v) ∩ Xv d(u, v′) + d(v, w) ≤ d(u, v′) + 2s ≤ d(u, v) + s d(v, v′) + d(u, w) ≤ s + d(u, v)

u v ′ v w s ≤ 2s ≤ d(u, v)

23/02/2017 Cop and Robber Game and Hyperbolicity 10/17

slide-32
SLIDE 32

δ-hyperbolic graphs are (2s, s + 2δ)-cop-win

Proposition (from Chepoi, Estellon ’07)

Any δ-hyperbolic graph is (2s, s + 2δ)∗-dismantlable, and thus (2s, s + 2δ)-cop-win Let w ∈ B2s(v) ∩ Xv d(u, v′) + d(v, w) ≤ d(u, v′) + 2s ≤ d(u, v) + s d(v, v′) + d(u, w) ≤ s + d(u, v) Consequently, d(v′, w) + d(u, v) ≤ s + d(u, v) + 2δ d(v′, w) ≤ s + 2δ

u v ′ v w s ≤ 2s ≤ d(u, v)

23/02/2017 Cop and Robber Game and Hyperbolicity 10/17

slide-33
SLIDE 33

δ-hyperbolic graphs are (2s, s + 2δ)-cop-win

Proposition (from Chepoi, Estellon ’07)

Any δ-hyperbolic graph is (2s, s + 2δ)∗-dismantlable, and thus (2s, s + 2δ)-cop-win

Question

Is any (s, s′)-cop-win graph f(s)-hyperbolic (when s′ < s) ?

23/02/2017 Cop and Robber Game and Hyperbolicity 10/17

slide-34
SLIDE 34

δ-hyperbolic graphs are (2s, s + 2δ)-cop-win

Proposition (from Chepoi, Estellon ’07)

Any δ-hyperbolic graph is (2s, s + 2δ)∗-dismantlable, and thus (2s, s + 2δ)-cop-win

Question

Is any (s, s′)-cop-win graph f(s)-hyperbolic (when s′ < s) ?

Theorem

G is (s, s − 1)-cop-win = ⇒ G is 64s2-hyperbolic

23/02/2017 Cop and Robber Game and Hyperbolicity 10/17

slide-35
SLIDE 35

Another characterization of hyperbolicity

For a cycle c, (D, Φ) is an N-filling of c if

◮ D is a 2-connected planar graph ◮ every internal face of D has at most

2N edges

◮ Φ : D → G is a simplicial map ◮ Φ(∂D) = c

c G D Φ

23/02/2017 Cop and Robber Game and Hyperbolicity 11/17

slide-36
SLIDE 36

Another characterization of hyperbolicity

For a cycle c, (D, Φ) is an N-filling of c if

◮ D is a 2-connected planar graph ◮ every internal face of D has at most

2N edges

◮ Φ : D → G is a simplicial map ◮ Φ(∂D) = c ◮ The area of (D, Φ) is the number of

faces of D

◮ AreaN(c) is the minimum area of an

N-filling of c

◮ ℓ(c) is the length of c

c G D Φ

23/02/2017 Cop and Robber Game and Hyperbolicity 11/17

slide-37
SLIDE 37

Another characterization of hyperbolicity

For a cycle c, (D, Φ) is an N-filling of c if

◮ D is a 2-connected planar graph ◮ every internal face of D has at most

2N edges

◮ Φ : D → G is a simplicial map ◮ Φ(∂D) = c ◮ The area of (D, Φ) is the number of

faces of D

◮ AreaN(c) is the minimum area of an

N-filling of c

◮ ℓ(c) is the length of c

1 2 3 4 5 6 7 8 9 10 11 12 13 14 1 5 2 6 14 9 13 10 12 4 7 11 11 7 4 23/02/2017 Cop and Robber Game and Hyperbolicity 11/17

slide-38
SLIDE 38

Another characterization of hyperbolicity

For a cycle c, (D, Φ) is an N-filling of c if

◮ D is a 2-connected planar graph ◮ every internal face of D has at most

2N edges

◮ Φ : D → G is a simplicial map ◮ Φ(∂D) = c ◮ The area of (D, Φ) is the number of

faces of D

◮ AreaN(c) is the minimum area of an

N-filling of c

◮ ℓ(c) is the length of c

1 2 3 4 5 6 7 8 9 10 11 12 13 14 1 5 2 6 14 9 13 8 23/02/2017 Cop and Robber Game and Hyperbolicity 11/17

slide-39
SLIDE 39

Linear Isoperimetric Inequality

A graph G satisfies the linear isoperimetric inequality, if there exists K ∈ N and N such that ∀c, AreaN(c) ≤ Kℓ(c)

Theorem (Gromov)

◮ G is δ-hyperbolic =

⇒ ∀c, Area16δ(c) ≤ ℓ(c)

◮ ∀c, AreaN(c) ≤ Kℓ(c) =

⇒ G is O(K 2N3)-hyperbolic For a proof, see [Bridson and Haefliger]

23/02/2017 Cop and Robber Game and Hyperbolicity 12/17

slide-40
SLIDE 40

Linear Isoperimetric Inequality

A graph G satisfies the linear isoperimetric inequality, if there exists K ∈ N and N such that ∀c, AreaN(c) ≤ Kℓ(c)

Theorem (Gromov)

◮ G is δ-hyperbolic =

⇒ ∀c, Area16δ(c) ≤ ℓ(c)

◮ ∀c, AreaN(c) ≤ Kℓ(c) =

⇒ G is O(K 2N3)-hyperbolic For a proof, see [Bridson and Haefliger]

Proposition

When K ∈ Q, ∀c, AreaN(c) ≤ ⌈Kℓ(c)⌉ = ⇒ G is (32KN2 + 1

2)-hyperbolic

23/02/2017 Cop and Robber Game and Hyperbolicity 12/17

slide-41
SLIDE 41

(s, s′)∗-dismantl. = ⇒ lin. isoperimetric inequality

Theorem

If G is (s, s′)∗-dismantlable with s′ < s, ∀c, Areas+s′(c) ≤

  • ℓ(c)

2(s − s′)

  • 23/02/2017

Cop and Robber Game and Hyperbolicity 13/17

slide-42
SLIDE 42

(s, s′)∗-dismantl. = ⇒ lin. isoperimetric inequality

Theorem

If G is (s, s′)∗-dismantlable with s′ < s, ∀c, Areas+s′(c) ≤

  • ℓ(c)

2(s − s′)

  • Proof by induction on ℓ(c):

◮ v: the last vertex of c in the

dismantling order

◮ Bs(v) ∩ c ⊆ Bs(v) ∩ Xv ⊆ Bs′(u)

c G v u x y s s ≤ s′ ≤ s′

23/02/2017 Cop and Robber Game and Hyperbolicity 13/17

slide-43
SLIDE 43

(s, s′)∗-dismantl. = ⇒ lin. isoperimetric inequality

Theorem

If G is (s, s′)∗-dismantlable with s′ < s, ∀c, Areas+s′(c) ≤

  • ℓ(c)

2(s − s′)

  • Proof by induction on ℓ(c):

◮ v: the last vertex of c in the

dismantling order

◮ Bs(v) ∩ c ⊆ Bs(v) ∩ Xv ⊆ Bs′(u) ◮ ℓ(c0) ≤ 2(s + s′) ◮ ℓ(c1) ≤ ℓ(c) − 2(s − s′)

G u x y c1 c0

23/02/2017 Cop and Robber Game and Hyperbolicity 13/17

slide-44
SLIDE 44

(s, s′)∗-dismantl. = ⇒ lin. isoperimetric inequality

Theorem

If G is (s, s′)∗-dismantlable with s′ < s, ∀c, Areas+s′(c) ≤

  • ℓ(c)

2(s − s′)

  • Proof by induction on ℓ(c):

◮ v: the last vertex of c in the

dismantling order

◮ Bs(v) ∩ c ⊆ Bs(v) ∩ Xv ⊆ Bs′(u) ◮ ℓ(c0) ≤ 2(s + s′) ◮ ℓ(c1) ≤ ℓ(c) − 2(s − s′)

G D u x y c1 c0 x y

23/02/2017 Cop and Robber Game and Hyperbolicity 13/17

slide-45
SLIDE 45

(s, s′)∗-dismantl. = ⇒ lin. isoperimetric inequality

Theorem

If G is (s, s′)∗-dismantlable with s′ < s, ∀c, Areas+s′(c) ≤

  • ℓ(c)

2(s − s′)

  • Proof by induction on ℓ(c):

◮ v: the last vertex of c in the

dismantling order

◮ Bs(v) ∩ c ⊆ Bs(v) ∩ Xv ⊆ Bs′(u) ◮ ℓ(c0) ≤ 2(s + s′) ◮ ℓ(c1) ≤ ℓ(c) − 2(s − s′) ◮ Areas+s′(c) ≤ 1+

  • ℓ(c1)

2(s−s′)

  • ℓ(c)

2(s−s′)

  • G

D u x y c1 c0 x y

23/02/2017 Cop and Robber Game and Hyperbolicity 13/17

slide-46
SLIDE 46

(s, s′)-cop-win graphs are hyperbolic

Theorem

G is (s, s′)∗-dismantlable with s′ < s = ⇒ δ∗(G) ≤ 16 (s+s′)2

s−s′

+ 1

2

Corollary

G is (s, s − 1)-cop-win = ⇒ G is 64s2-hyperbolic

23/02/2017 Cop and Robber Game and Hyperbolicity 14/17

slide-47
SLIDE 47

Computing the hyperbolicity

Assume the distance-matrix of G has been computed Computing the hyperbolicity δ∗(G)

◮ 4 points condition: O(n4)

Computing an approximation of δ∗(G)

◮ fixing one point: a 2-approx. in O(n3)

23/02/2017 Cop and Robber Game and Hyperbolicity 15/17

slide-48
SLIDE 48

Computing the hyperbolicity

Assume the distance-matrix of G has been computed Computing the hyperbolicity δ∗(G)

◮ 4 points condition: O(n4) ◮ Using (max, min)-matrix product: O(n3.69)

[Fournier, Ismail, Vigneron ’12] Computing an approximation of δ∗(G)

◮ fixing one point: a 2-approx. in O(n3) ◮ Using (max, min)-matrix product: a 2-approx. in O(n2.69)

[Fournier, Ismail, Vigneron ’12]

◮ a (2 + ǫ)-approx. in O( 1 ǫ n2.38)

[Duan ’14]

23/02/2017 Cop and Robber Game and Hyperbolicity 15/17

slide-49
SLIDE 49

Computing the hyperbolicity

Assume the distance-matrix of G has been computed Computing the hyperbolicity δ∗(G)

◮ 4 points condition: O(n4) ◮ Using (max, min)-matrix product: O(n3.69)

[Fournier, Ismail, Vigneron ’12] Computing an approximation of δ∗(G)

◮ fixing one point: a 2-approx. in O(n3) ◮ Using (max, min)-matrix product: a 2-approx. in O(n2.69)

[Fournier, Ismail, Vigneron ’12]

◮ a (2 + ǫ)-approx. in O( 1 ǫ n2.38)

[Duan ’14]

Theorem

From the distance-matrix of G, one can compute a constant approximation of δ∗(G) in O(n2)

23/02/2017 Cop and Robber Game and Hyperbolicity 15/17

slide-50
SLIDE 50

Approximation Algorithm for δ∗

Approx-δ∗(G,α) Consider a BFS ordering ≺ of V(G) from any vertex u ; For all v, let fα(v) be on a shortest path from v to u such that d(v, fα(v)) = 2α ; for all v ∈ V do if B4α(v, G) ∩ Xv ⊆ B3α(fα(v), G) then return NO return YES;

u fα(v) v 2α Xv

23/02/2017 Cop and Robber Game and Hyperbolicity 16/17

slide-51
SLIDE 51

Approximation Algorithm for δ∗

Approx-δ∗(G,α) Consider a BFS ordering ≺ of V(G) from any vertex u ; For all v, let fα(v) be on a shortest path from v to u such that d(v, fα(v)) = 2α ; for all v ∈ V do if B4α(v, G) ∩ Xv ⊆ B3α(fα(v), G) then return NO return YES;

u fα(v) v 2α Xv

NO ≺ is not (2(2α), 2α + α)∗-dismantling = ⇒ δ∗ > α

2

YES G is (4α, 3α)∗-dismantlable = ⇒ δ∗ ≤ 16(7α)2

α

+ 1

2 = 784α + 1 2

23/02/2017 Cop and Robber Game and Hyperbolicity 16/17

slide-52
SLIDE 52

Approximation Algorithm for δ∗

Approx-δ∗(G,α) Consider a BFS ordering ≺ of V(G) from any vertex u ; For all v, let fα(v) be on a shortest path from v to u such that d(v, fα(v)) = 2α ; for all v ∈ V do if B4α(v, G) ∩ Xv ⊆ B3α(fα(v), G) then return NO return YES;

u fα(v) v 2α Xv

NO ≺ is not (2(2α), 2α + α)∗-dismantling = ⇒ δ∗ > α

2

YES G is (4α, 3α)∗-dismantlable = ⇒ δ∗ ≤ 16(7α)2

α

+ 1

2 = 784α + 1 2

We can find α∗ α∗/2 ≤ δ∗ ≤ 784α∗ + 1

2

1569-approx. of δ∗(G)

23/02/2017 Cop and Robber Game and Hyperbolicity 16/17

slide-53
SLIDE 53

Approximation Algorithm for δ∗

Approx-δ∗(G,α) Consider a BFS ordering ≺ of V(G) from any vertex u ; For all v, let fα(v) be on a shortest path from v to u such that d(v, fα(v)) = 2α ; for all v ∈ V do if B4α(v, G) ∩ Xv ⊆ B3α(fα(v), G) then return NO return YES;

u fα(v) v 2α Xv

Complexity: Approx-δ∗(G,α) runs in time O(n2)

Proposition

One can compute a 1569-approximation of δ∗ in time O(n2 log δ∗)

23/02/2017 Cop and Robber Game and Hyperbolicity 16/17

slide-54
SLIDE 54

Approximation Algorithm for δ∗

Approx-δ∗(G,α) Consider a BFS ordering ≺ of V(G) from any vertex u ; For all v, let fα(v) be on a shortest path from v to u such that d(v, fα(v)) = 2α ; for all v ∈ V do if B4α(v, G) ∩ Xv ⊆ B3α(fα(v), G) then return NO return YES;

u fα(v) v 2α Xv

We can avoid to recompute everything when we increase α

Theorem

One can compute a 1569-approximation of δ∗ in time O(n2)

23/02/2017 Cop and Robber Game and Hyperbolicity 16/17

slide-55
SLIDE 55

Conclusion

◮ Characterization of hyperbolicity via a cop and robber

game Different notions that are qualitatively equivalent

◮ (s, s′)-cop-win graphs ◮ (s, s′)-dismantlability ◮ (s, s′)∗-dismantlability ◮ bounded hyperbolicity

◮ Links between (s, s′)∗-dismantlability and hyperbolicity

hold for infinite graphs

◮ A constant-factor approximation of the hyperbolicity in

O(n2) (starting from the distance-matrix)

23/02/2017 Cop and Robber Game and Hyperbolicity 17/17