Slides
-
Upload
sergey-kirgizov -
Category
Technology
-
view
58 -
download
0
Transcript of Slides
![Page 1: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/1.jpg)
On the complexity of turning a graph intothe analogue of a clique
Julien Bensmail 1 Romaric Duvignau 1
Sergey Kirgizov 2
1Combinatoire et Algorithmique ⊂ LaBRI ⊂ (Université de Bordeaux × CNRS)
2Complex Networks ⊂ LIP6 ⊂ (UPMC × CNRS)
30 June – 4 July2014
![Page 2: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/2.jpg)
Gorientation7−−−−−−−→ −→G
![Page 3: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/3.jpg)
Sometimes−→G is o-clique
![Page 4: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/4.jpg)
Does such orientation existfor a given graph?
![Page 5: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/5.jpg)
How difficult is it to findthe answer?
![Page 6: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/6.jpg)
Outline
1 Some definitions
2 Can a given graph be turned into o-clique?
3 Family of related problems
4 Open question
3
![Page 7: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/7.jpg)
Definitions:orientations, o-cliques, oriented distances and diameters,Klostermeyer-MacGillivray lemma
![Page 8: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/8.jpg)
Orientation
Gorientation7−−−−−−→ −→G is an assignment of a direction to each edge
from undirected graph.
••
•
•
••
•
•
••
•
•
o1
o2
4
![Page 9: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/9.jpg)
Oriented chromatic number
χo(−→G ) is the minimal number of colors such that:
1 Colors of adjacent vertices are different
• •
2 All arcs between two colors have the same direction• •
• •
• •
5
![Page 10: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/10.jpg)
O-cliqueoriented analogue of usual clique
Undirected
G is clique ⇐⇒ χ(G ) = n
•
•
••
Oriented−→G is o-clique ⇐⇒ χo (−→G ) = n
• •
••
• • •
6
![Page 11: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/11.jpg)
Oriented distances and diameters
Strong
dists = max
(dist(u, v )dist(v , u)
)
diams = maxu,v∈−→G
dists (u, v )
Weak
distw = min
(dist(u, v )dist(v , u)
)
diamw = maxu,v∈−→G
distw (u, v )
7
![Page 12: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/12.jpg)
O-cliques & weak diameter
−→G is o-clique
χo (−→G ) = n diamw (−→G ) ≤ 2
[Klostermeyer, MacGillivray, 2004]
8
![Page 13: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/13.jpg)
Decision problem:Does G admit a 2-weak orientation?
![Page 14: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/14.jpg)
Does G admit a 2-weak orientation?
An orientation G 7→ −→G is called 2-weak when diamw (−→G ) ≤ 2
• • • ?
9
![Page 15: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/15.jpg)
Does G admit a 2-weak orientation?
An orientation G 7→ −→G is called 2-weak when diamw (−→G ) ≤ 2
• • • • • •
9
![Page 16: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/16.jpg)
Does G admit a 2-weak orientation?
An orientation G 7→ −→G is called 2-weak when diamw (−→G ) ≤ 2
• • •
••
•
•
• • •
?
9
![Page 17: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/17.jpg)
Does G admit a 2-weak orientation?
An orientation G 7→ −→G is called 2-weak when diamw (−→G ) ≤ 2
• • •
••
•
•
• • •
••
•
•
••
•
•
••
•
•
••
•
•
9
![Page 18: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/18.jpg)
Does G admit a 2-weak orientation?
An orientation G 7→ −→G is called 2-weak when diamw (−→G ) ≤ 2
• • •
••
•
•
• • •
••
•
•
••
•
•
••
•
•
••
•
•
diamw =∞ for any orientation of a claw9
![Page 19: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/19.jpg)
2-weak orientation is NP-complete
2-weak orientation is in NP
We just run BFS from all vertices to check whetherdiamw (−→G ) ≤ 2 or not.
2-weak orientation is NP-hard
We prove this by reduction from the monotone versionof Not-All-Equal 3SAT
10
![Page 20: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/20.jpg)
2-weak orientation is NP-complete(xi ∨ xj ∨ xk ) ∧ . . . ∧ (xi ∨ xk ∨ xl ) 7→ G
Monotone Not-All-Equal 3SAT
F is 3CNF formula without negations.Example: (x1 ∨ x2 ∨ x3) ∧ (x3 ∨ x3 ∨ x4) ∧ . . .Is F satisfiable in such way that no clausehave all literals set to same value?
NP-complete even when in every clause all variablesare different
11
![Page 21: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/21.jpg)
2-weak orientation is NP-complete(xi ∨ xj ∨ xk ) ∧ . . . ∧ (xi ∨ xk ∨ xl ) 7→ G
Monotone Not-All-Equal 3SAT
F is 3CNF formula without negations.Example: (x1 ∨ x2 ∨ x3) ∧ (x3 ∨ x3 ∨ x4) ∧ . . .Is F satisfiable in such way that no clausehave all literals set to same value?
NP-complete even when in every clause all variablesare different
11
![Page 22: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/22.jpg)
2-weak orientation is NP-complete(xi ∨ xj ∨ xk ) ∧ . . . ∧ (xi ∨ xk ∨ xl ) 7→ G
Reduction overview:
F G
2this diagram commutes
F = nae-3-formulaeG = graphs2 = {good, bad}
12
![Page 23: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/23.jpg)
2-weak orientation is NP-complete(xi ∨ xj ∨ xk ) ∧ . . . ∧ (xi ∨ xk ∨ xl ) 7→ G
F 7→ G
F is nae-satisfiable ⇐⇒ ∃−→G : diamw (−→G ) ≤ 2
(•, •′) is a representative pair from G
(•, •′) is a non-representative pair from G
∃−→G : distw (•, •′) ≤ 2 ⇐⇒ F is nae-satisfiable
∃−→G : distw (•, •′) ≤ 2 by construction of G
13
![Page 24: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/24.jpg)
2-weak orientation is NP-complete(xi ∨ xj ∨ xk ) ∧ . . . ∧ (xi ∨ xk ∨ xl ) 7→ G
Variable gadget:
xi 7−−→ •vi •
v ′i
xi = TRUE 7−−→ •vi •
v ′i
xi = FALSE 7−−→ •vi •
v ′i
14
![Page 25: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/25.jpg)
2-weak orientation is NP-complete(xi ∨ xj ∨ xk ) ∧ . . . ∧ (xi ∨ xk ∨ xl ) 7→ G
Gadget for clause:
Ct = xi ∨ xj ∨ xk 7−−→
•vi •
v ′i •vi ,t
•vj
•v ′j
•vj ,t
•vk •
v ′k •vk,t
•wt
•w ′t
Ct is nae-satisfiable ⇐⇒ ∃−→G : distw (•, •′) ≤ 2
Using w ′t we can make distw (•, •′) ≤ 2when it is needed
15
![Page 26: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/26.jpg)
2-weak orientation is NP-complete(xi ∨ xj ∨ xk ) ∧ . . . ∧ (xi ∨ xk ∨ xl ) 7→ G
Gadget for clause:
Ct = xi ∨ xj ∨ xk 7−−→
•vi •
v ′i •vi ,t
•vj
•v ′j
•vj ,t
•vk •
v ′k •vk,t
•wt
•w ′t
Ct is nae-satisfiable ⇐⇒ ∃−→G : distw (•, •′) ≤ 2
Using w ′t we can make distw (•, •′) ≤ 2when it is needed
15
![Page 27: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/27.jpg)
2-weak orientation is NP-complete(xi ∨ xj ∨ xk ) ∧ . . . ∧ (xi ∨ xk ∨ xl ) 7→ G
Gadget for clause:
Ct = xi ∨ xj ∨ xk 7−−→
•vi •
v ′i •vi ,t
•vj
•v ′j
•vj ,t
•vk •
v ′k •vk,t
•wt
•w ′t
Ct is nae-satisfiable ⇐⇒ ∃−→G : distw (•, •′) ≤ 2
Using w ′t we can make distw (•, •′) ≤ 2when it is needed
15
![Page 28: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/28.jpg)
2-weak orientation is NP-complete(xi ∨ xj ∨ xk ) ∧ . . . ∧ (xi ∨ xk ∨ xl ) 7→ G
Gadget for clause:
Ct = xi ∨ xj ∨ xk 7−−→
•vi •
v ′i •vi ,t
•vj
•v ′j
•vj ,t
•vk •
v ′k •vk,t
•wt
•w ′t
Ct is nae-satisfiable ⇐⇒ ∃−→G : distw (•, •′) ≤ 2
Using w ′t we can make distw (•, •′) ≤ 2when it is needed
15
![Page 29: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/29.jpg)
2-weak orientation is NP-complete(xi ∨ xj ∨ xk ) ∧ . . . ∧ (xi ∨ xk ∨ xl ) 7→ G
Gadget for clause:
Ct = xi ∨ xj ∨ xk 7−−→
•vi •
v ′i •vi ,t
•vj
•v ′j
•vj ,t
•vk •
v ′k •vk,t
•wt
•w ′t
Ct is nae-satisfiable ⇐⇒ ∃−→G : distw (•, •′) ≤ 2
Using w ′t we can make distw (•, •′) ≤ 2when it is needed
15
![Page 30: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/30.jpg)
2-weak orientation is NP-complete(xi ∨ xj ∨ xk ) ∧ . . . ∧ (xi ∨ xk ∨ xl ) 7→ G
Gadget for clause:
Ct = xi ∨ xj ∨ xk 7−−→
•vi •
v ′i •vi ,t
•vj
•v ′j
•vj ,t
•vk •
v ′k •vk,t
•wt
•w ′t
Ct is nae-satisfiable ⇐⇒ ∃−→G : distw (•, •′) ≤ 2
Using w ′t we can make distw (•, •′) ≤ 2when it is needed
15
![Page 31: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/31.jpg)
2-weak orientation is NP-complete(xi ∨ xj ∨ xk ) ∧ . . . ∧ (xi ∨ xk ∨ xl ) 7→ G
Gadget for clause:
Ct = xi ∨ xj ∨ xk 7−−→
•vi •
v ′i •vi ,t
•vj
•v ′j
•vj ,t
•vk •
v ′k •vk,t
•wt
•w ′t
Ct is nae-satisfiable ⇐⇒ ∃−→G : distw (•, •′) ≤ 2
Using w ′t we can make distw (•, •′) ≤ 2when it is needed
15
![Page 32: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/32.jpg)
2-weak orientation is NP-complete(xi ∨ xj ∨ xk ) ∧ . . . ∧ (xi ∨ xk ∨ xl ) 7→ G
Gadget for clause:
Ct = xi ∨ xj ∨ xk 7−−→
•vi •
v ′i •vi ,t
•vj
•v ′j
•vj ,t
•vk •
v ′k •vk,t
•wt
•w ′t
Ct is nae-satisfiable ⇐⇒ ∃−→G : distw (•, •′) ≤ 2
Using w ′t we can make distw (•, •′) ≤ 2when it is needed
15
![Page 33: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/33.jpg)
2-weak orientation is NP-complete(xi ∨ xj ∨ xk ) ∧ . . . ∧ (xi ∨ xk ∨ xl ) 7→ G
Gadget for clause:
Ct = xi ∨ xj ∨ xk 7−−→
•vi •
v ′i •vi ,t
•vj
•v ′j
•vj ,t
•vk •
v ′k •vk,t
•wt
•w ′t
Ct is nae-satisfiable ⇐⇒ ∃−→G : distw (•, •′) ≤ 2
Using w ′t we can make distw (•, •′) ≤ 2when it is needed
15
![Page 34: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/34.jpg)
2-weak orientation is NP-complete(xi ∨ xj ∨ xk ) ∧ . . . ∧ (xi ∨ xk ∨ xl ) 7→ G
Gadget for clause:
Ct = xi ∨ xj ∨ xk 7−−→
•vi •
v ′i •vi ,t
•vj
•v ′j
•vj ,t
•vk •
v ′k •vk,t
•wt
•w ′t
Ct is nae-satisfiable ⇐⇒ ∃−→G : distw (•, •′) ≤ 2
Using w ′t we can make distw (•, •′) ≤ 2when it is needed
15
![Page 35: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/35.jpg)
2-weak orientation is NP-complete(xi ∨ xj ∨ xk ) ∧ . . . ∧ (xi ∨ xk ∨ xl ) 7→ G
Gadget for clause:
Ct = xi ∨ xj ∨ xk 7−−→
•vi •
v ′i •vi ,t
•vj
•v ′j
•vj ,t
•vk •
v ′k •vk,t
•wt
•w ′t
Ct is nae-satisfiable ⇐⇒ ∃−→G : distw (•, •′) ≤ 2
Using w ′t we can make distw (•, •′) ≤ 2when it is needed
15
![Page 36: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/36.jpg)
2-weak orientation is NP-complete(xi ∨ xj ∨ xk ) ∧ . . . ∧ (xi ∨ xk ∨ xl ) 7→ G
GF is a union of clause gadgets. G = GF + smth.For any pair of vertices (u, v ) from GF we add the following:
•u
• •
• •
•v
distw (u, •) ≤ 2distw (v , •) ≤ 2
Here distw (u, v ) = 3, so we don’t break ourreduction by adding blue vertices.
16
![Page 37: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/37.jpg)
2-weak orientation is NP-complete(xi ∨ xj ∨ xk ) ∧ . . . ∧ (xi ∨ xk ∨ xl ) 7→ G
Also, for any non-representative pair of vertices (u,v) from GF
we add the following:
•u
• •
• •
•v
•
distw (u, •) = 1distw (v , •) = 1
Here distw (u, v ) = 2, so we ensure that the distancebetween non-representative vertices is ≤ 2
17
![Page 38: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/38.jpg)
2-weak orientation is NP-complete(xi ∨ xj ∨ xk ) ∧ . . . ∧ (xi ∨ xk ∨ xl ) 7→ G
Next, in order to ensure that ∀y ∈ GF : distw (y , •) ≤ 2we add the following:
•u
• •
• •
•v
••
• y — any other vertex from GF
•
distw (y , •) = 2
18
![Page 39: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/39.jpg)
2-weak orientation is NP-complete(xi ∨ xj ∨ xk ) ∧ . . . ∧ (xi ∨ xk ∨ xl ) 7→ G
Finally, turn all brown and blue vertices into one big clique,by adding all possible edges between them.
19
![Page 40: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/40.jpg)
2-weak orientation is NP-complete(xi ∨ xj ∨ xk ) ∧ . . . ∧ (xi ∨ xk ∨ xl ) 7→ G
∃−→G : distw (•, •′) ≤ 2 ⇐⇒ F is nae-satisfiable
∃−→G : distw (•, •′) ≤ 2
∃−→G : distw (•, •′) ≤ 2
∃−→G : distw (•, •′) ≤ 2
∃−→G : distw (•, •′) ≤ 2
∃−→G : distw (•, •′) ≤ 2
∃−→G : distw (•, •′) ≤ 2
by construction of G
F is nae-satisfiable ⇐⇒ ∃−→G : diamw (−→G ) ≤ 2
20
![Page 41: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/41.jpg)
Family of related problems:Does G admit an orientation G 7→ −→G such that prop(−→G )?
![Page 42: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/42.jpg)
Family of o-problemsDoes G admit an orientation G 7→ −→G such that prop(−→G )?
Strong orientation:
diams (−→G ) <∞ ⇐⇒ G is bridgeless [Robbins, 1939]
Linear time using Tarjan’s bridge-finding algorithm [1974]
k-strong orientation:
diams (−→G ) ≤ k
NP-complete for k = 2 [Chvátal, Thomassen, 1978]
21
![Page 43: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/43.jpg)
Family of o-problemsDoes G admit an orientation G 7→ −→G such that prop(−→G )?
Weak orientation:
diamw (−→G ) <∞ ⇐⇒ B-contraction(G ) is claw-free.
B-contraction means “replace allbridgeless subgraphs by vertices”
Linear time [Bensmail, Duvignau, Kirgizov 2013]
22
![Page 44: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/44.jpg)
Family of o-problemsDoes G admit an orientation G 7→ −→G such that prop(−→G )?
k-weak orientation:
diamw (−→G ) ≤ k
NP-complete for k ≥ 2 [Bensmail, Duvignau, Kirgizov, 2013]
Gk is a k-edge-coloured graph(e.g. 2-edge coloured graphs are just signified graphs)
∃−→G : diamw (−→G ) ≤ k ⇐⇒ ∃Gk : χk (Gk ) = n
23
![Page 45: Slides](https://reader033.fdocuments.us/reader033/viewer/2022042821/55d6cb87bb61ebe8188b459e/html5/thumbnails/45.jpg)
Open question:
Complexity of k-strongorientation when k > 2
Thank you for your attention
kirgizov.complexnetworks.fr/some-problems/orientation.html