Weight Choosability of theta Graphs
Weight Choosability of theta Graphs
Ting-Feng Jian
Advisor: Gerard Jennhwa Chang
Department of Mathematics, National Taiwan University
August 2, 2014
Weight Choosability of theta Graphs
Outline
1. Introduction 2. The Paths 3. The Cycles 4. The θ-graphs
5. The Generalized θ-graphs 6. Further Problems
Weight Choosability of theta Graphs Introduction
Definitions
1. G = (V , E ) be a connected graph but not K2.
Weight Choosability of theta Graphs Introduction
Definitions
2. L(e) ⊆ R, a list of weights of e.
Weight Choosability of theta Graphs Introduction
Definitions
3. L-edge weighting: f such that f (e) ∈ L(e).
Weight Choosability of theta Graphs Introduction
Definitions
4. induced weight: g (v ) =P
uv∈Ef(uv ).
Weight Choosability of theta Graphs Introduction
Definitions
5. proper weighting: g (v ) 6= g (v′).
Weight Choosability of theta Graphs Introduction
Definitions
1. L(e) = {1, 2, ..., k}, f proper weighting
⇒ G is k-edge weight colorable.
2. L(e) ⊆ R, f proper weighting
⇒ G is k-edge weight choosable.
Weight Choosability of theta Graphs Introduction
Problems
Conjecture (M.Karonski, T.Luczak, and A.Thomason) ([1]) Every connected graph G 6= K2 is 3-edge weight colorable.
Conjecture (T.Bartnicki, J.Grytczuk, and S.Niwczykl) ([2]) Every connected graph G 6= K2 is 3-edge weight choosable.
Weight Choosability of theta Graphs Introduction
Recent Result
Theorem (M.Kalkowski, M.Karonski, and F.Pfender, 2010) ([8]) Every connected graph G 6= K2 is 5-edge weight colorable.
Theorem (T.Bartnicki, J.Grytczuk, and S.Niwczyk1)
([2]) A clique, complete bipartite graph, or a tree, not K2, is 3-edge weight choosable.
Weight Choosability of theta Graphs Introduction
Polynomials
1. Edge Set: E = {e1, e2, ..., em}.
2. Variables xe = f (e) ∈ L(e).
3. Associated polynomial of G of orientation D:
PG(x1, x2, ..., xm) = Y
vv′∈E (D)
X
e=uv ∈E
xe− X
e′=u′v′∈E
xe′
! 6= 0
Weight Choosability of theta Graphs Introduction
Example
Then the associated polynomial: PG(x1, x2, x3, x4, x5)
= (x4−x2−x5)(x1+x5−x3)(x2−x4−x5)(x3+x5−x1)(x1+x2−x3−x4).
Weight Choosability of theta Graphs Introduction
Combinatorial Nullstellensatz
Theorem (N. Alon, 1999)
([3]) Let F be an arbitrary field, and let P(x1, x2, ..., xm) be a polynomial in F[x1, x2, ..., xm]. Suppose that the coefficient of x1k1x2k2...xmkm in P is non-zero and deg(P) =Pm
i=1ki. Then for any subsets A1, A2, ..., Am of F satisfying |Ai| ≥ ki + 1 for all i = 1, 2, ..., m, there exists
(a1, a2, ..., am) ∈ A1 × A2× ... × An so that P(a1, a2, ..., am) 6= 0.
Weight Choosability of theta Graphs Introduction
Monomial Index
Define the monomial index by mind(P) = min
M h(M) = min
M max
1≤i ≤mki. Coeffiecient of x1x2...xm is non-zero
⇒ 2-egde Weight Choosable.
Coeffiecient of x1k1x2k2...xmk2 is non-zero for ki ≤ 2
⇒ 3-egde Weight Choosable.
Weight Choosability of theta Graphs Introduction
Example
Then the associated polynomial: PG(x1, x2, x3, x4, x5)
= (x4−x2−x5)(x1+x5−x3)(x2−x4−x5)(x3+x5−x1)(x1+x2−x3−x4).
Weight Choosability of theta Graphs Introduction
Permanent
Let m × m matrix A = [aij].
1. Permanent:
perA = X
σ∈Sm
m
Y
i=1
ai σ(i )
! .
2. Let K = (k1, k2, ..., km), ki ≥ 0 and Pm
i=1ki = m.
Repeating the i -th columns ki times, denoted A(K ).
Weight Choosability of theta Graphs Introduction
Permanent Index
permanent index:
The minimum of k so that there is
K = (k1, k2, ..., km), ki ≤ k for all i and perA(K ) 6= 0.
Weight Choosability of theta Graphs Introduction
Orientation
Fixed orientation D of a graph G , define the associated matrix AG = [aij] by
aij =
1, if ej is incident to the head of ei;
−1, if ej is incident to the tail of ei; 0, if ej and ei are not incident.
Weight Choosability of theta Graphs Introduction
The Relation
PG(x1, x2, x3, x4, x5) AG
= (x4− x2− x5) 0 −1 0 1 −1 (x1+ x5− x3) 1 0 −1 0 1 (x2− x4− x5) 0 1 0 −1 −1 (x3+ x5− x1) −1 0 1 0 1 (x1+ x2− x3− x4) 1 1 −1 −1 0
Weight Choosability of theta Graphs Introduction
The Relation
Coefficient of x1x2x3x4x5: perAG.
Coefficient of x12x2x3x4: perAG(2, 1, 1, 1, 0)/2!.
Coefficient of x12x22x3: perAG(2, 2, 1, 0, 0)/2!2!.
Coefficient of x13x2x3: perAG(3, 1, 1, 0, 0)/3!.
Weight Choosability of theta Graphs Introduction
The Lemma
Lemma
([2]) Let A = [aij] be a m × m matrix with finite permanent index. Let the polynomial
P(x1, x2, ..., xm) =
m
Y
i=1 m
X
j=1
aijxj
! ,
then mind(P) = pind(A).
Weight Choosability of theta Graphs Introduction
Useful Result
Theorem
([2]) Let AG be an associated matrix of G . If pind(AG) ≤ k, then G is (k + 1)-edge weight choosable.
Weight Choosability of theta Graphs The Paths
Paths
Let path Pm : v1v2...vm+1 with m edges.
Weight Choosability of theta Graphs The Paths
Associated Matrices
APm =
0 −1 0 0 0 . ..
1 0 −1 0 . .. 0
0 1 0 . .. 0 0
0 0 . .. 0 −1 0
0 . .. 0 1 0 −1
. .. 0 0 0 1 0
.
Weight Choosability of theta Graphs The Paths
2-Choosability
Theorem
Let Pm be a path. Let APm be the associated matrix of Pm, m ≥ 2. Then
perAPm = (−1)m2, if m is even 0, otherwise.
AP2 =0 −1 1 0
AP3 =
0 −1 0
1 0 −1
0 1 0
Weight Choosability of theta Graphs The Paths
3-Choosability
Lemma
Let APm be the associated matrix of path Pm with m ≥ 4 edges. Let K = (k1, k2, ..., km) where k1 = km = 0,
k2 = k3 = 2, and other ki = 1. Then perAPm(K ) = 4 K = (0, 2, 2, 1, ..., 1, 0)
Weight Choosability of theta Graphs The Paths
3-Choosability
APm(K ) =
−1 −1 0 0 0 0 · · · 0 0
0 0 −1 −1 0 0 · · · 0 0
1 1 0 0 −1 0 · · · 0 0
0 0 1 1 0 −1 · · · 0 0
0 0 0 0 1 0 · · · 0 0
... ... ... ... ... ... ... ... ...
0 0 0 0 0 0 · · · 0 −1
0 0 0 0 0 0 · · · 1 0
0 0 0 0 0 0 · · · 0 1
,
Weight Choosability of theta Graphs The Paths
Non-2-Choosability
P3 is 2-choosable. Take x1, x3 so that x1 6= x3 and x2 6= 0.
Weight Choosability of theta Graphs The Paths
Non-2-Choosability
Pm is not 2-choosable for odd m ≥ 5.
(1) xi 6= xi+2 and x2 6= 0, xm−16= 0.
(2) Assign L(e2j) = {j − 1, j} for j = 1, 2, ...,m−32
Weight Choosability of theta Graphs The Cycles
Cycles
Let E = {e1, e2, ..., en} be the edge set of Cn. Give the orientation as ei+1 follows ei for i = 1, 2, ..., n − 1 and e1
follows en.
Weight Choosability of theta Graphs The Cycles
Associated Matrices
ACn =
0 −1 0 0 · · · 0 0 1
1 0 −1 0 · · · 0 0 0
0 1 0 −1 · · · 0 0 0
0 0 1 0 · · · 0 0 0
... ... ... ... . .. ... ... ...
0 0 0 0 · · · 0 −1 0
0 0 0 0 · · · 1 0 −1
−1 0 0 0 · · · 0 1 0
.
Weight Choosability of theta Graphs The Cycles
2-Choosability
n is odd: an= 1n+ (−1)n = 0.
n is even:
bij =
1, if i − j = 0;
−1, if i − j = −1(mod n);
0, otherwise.
an= (1n2 + (−1)n2)bn2 = 4, if 4 divides n 0, otherwise.
Weight Choosability of theta Graphs The Cycles
2-Choosability
Theorem
Let Cn be a cycle. Let ACn be the associated matrix of Cn, n ≥ 3. Then
perACn = 4, if 4 divides n;
0, otherwise.
Weight Choosability of theta Graphs The Cycles
3-Choosability
Theorem
Let ACn be the associated matrix of Cn, n ≥ 4.
Let K = (k1, k2, ..., kn) where k1 = k2 = 2, k3 = k4 = 0 and other ki = 1.
Then
perACn(K ) = (−1)n× 4.
In particular, perACn(K ) 6= 0.
Weight Choosability of theta Graphs The Cycles
Associated Matrices
ACn =
0 −1 0 0 · · · 0 0 1
1 0 −1 0 · · · 0 0 0
0 1 0 −1 · · · 0 0 0
0 0 1 0 · · · 0 0 0
... ... ... ... . .. ... ... ...
0 0 0 0 · · · 0 −1 0
0 0 0 0 · · · 1 0 −1
−1 0 0 0 · · · 0 1 0
.
Weight Choosability of theta Graphs The Cycles
Finding A(K )
AC4(K ) =
0 0 −1 −1
1 1 0 0
0 0 1 1
−1 −1 0 0
AC5(K ) =
0 0 −1 −1 1
1 1 0 0 0
0 0 1 1 0
0 0 0 0 −1
−1 −1 0 0 0
a4 = 4, a5 = −4.
Weight Choosability of theta Graphs The Cycles
Finding A(K )
ACn(K ) =
0 0 −1 −1 0 · · · 0 1
1 1 0 0 0 · · · 0 0
0 0 1 1 0 · · · 0 0
0 0 0 0 −1 · · · 0 0
0 0 0 0 0 · · · 0 0
... ... ... ... ... . .. ... ...
0 0 0 0 0 · · · 0 −1
−1 −1 0 0 0 · · · 1 0
.
an = an−2 = (−1)n× 4.
Weight Choosability of theta Graphs The Cycles
Non-2-Choosability
Theorem
If 4 does not divide n, then Cn is not 2-edge weight colorable.
Weight Choosability of theta Graphs The Cycles
Consequences
2-edge weight choosable:
P3, Pm for even m and Cn for 4 divides n.
3-edge weight choosable:
Pm for odd m 6= 3 and Cn for 4 does not divide n.
Weight Choosability of theta Graphs The θ-graphs
What Is a θ-graph?
θ(m1, m2, m3) for the θ-graph if the lengths of the upper, middle, and lower paths are m1, m2, m3, respectively.
Weight Choosability of theta Graphs The θ-graphs
Associated Matrices
Aθ(3,2,3) =
0 −1 0 1 0 1 0 0
1 0 −1 0 0 0 0 0
0 1 0 0 −1 0 0 −1
1 0 0 0 −1 1 0 0
0 0 −1 1 0 0 0 −1
1 0 0 1 0 0 −1 0
0 0 0 0 0 1 0 −1
0 0 −1 0 −1 0 1 0
.
Weight Choosability of theta Graphs The θ-graphs
Associated Matrices
AX AXY AXZ AYX AY AYZ AZX AZY AZ
,
AX, AY, and AZ: associated matrix of paths of lengths m1, m2, m3, respectively.
Other submatrices have only two numbers: 1 on the upper left and −1 on the lower right.
Weight Choosability of theta Graphs The θ-graphs
Notations
1. S = (R, C ) where |R| = |C | and
R ⊆ {1, 2, ..., m}, C ⊆ {1, 2, ..., m}.
2. AS: submatrix of A formed by the R rows and C -th columns.
3. A(S): submatrix of A obtained by deleting the R rows and C columns.
Weight Choosability of theta Graphs The θ-graphs
The Main Proposition
Proposition
Let Aθ(m1,m2,m3) be the associated matrix of θ(m1, m2, m3).
Let S3 demote the permutation group of rank 3. Then 1. perAθ(m1,m2,m3)= perAθ(mσ(1),mσ(2),mσ(3)) for all σ ∈ S3. 2. perAθ(m1+4,m2,m3)= perAθ(m1,m2,m3) for m1 ≥ 3.
Weight Choosability of theta Graphs The θ-graphs
The Proof
Let A = Aθ(m1+4,m2,m3) and B = Aθ(m1,m2,m3). Such A has the following form:
A=
0 −1 0 0 0 0 0 · · ·
1 0 −1 0 0 0 0 · · ·
0 1 0 −1 0 0 0 · · ·
0 0 1 0 −1 0 0 · · · AXY AXZ
0 0 0 1 0 −1 0 · · ·
0 0 0 0 1 0 −1 · · ·
0 0 0 0 0 1 0 · · ·
... ... ... ... ... ... ... . .. ... ...
AYX AY AYZ
AZX AZY AZ
.
Choose R = {2, 3, 4, 5, 6} with |R| = 5.
Weight Choosability of theta Graphs The θ-graphs
The Proof
perAS1 = perA(R,{1,2,3,4,5}) = per
1 0 −1 0 0
0 1 0 −1 0
0 0 1 0 −1
0 0 0 1 0
0 0 0 0 1
= 1.
perAS2 = perA(R,{1,3,4,5,6}) = per
1 −1 0 0 0
0 0 −1 0 0
0 1 0 −1 0
0 0 1 0 −1
0 0 0 1 0
= 1.
Weight Choosability of theta Graphs The θ-graphs
The Proof
perAS3 = perA(R,{1,2,3,4,7}) = per
1 0 −1 0 0
0 1 0 −1 0
0 0 1 0 0
0 0 0 1 0
0 0 0 0 −1
= 0.
perAS4 = perA(R,{2,3,4,5,6}) = per
0 −1 0 0 0
1 0 −1 0 0
0 1 0 −1 0
0 0 1 0 −1
0 0 0 1 0
= 0.
Weight Choosability of theta Graphs The θ-graphs
The Proof
perAS5 = perA(R,{2,3,4,5,7}) = per
0 −1 0 0 0
1 0 −1 0 0
0 1 0 −1 0
0 0 1 0 0
0 0 0 1 −1
= −1.
perAS6 = perA(R,{3,4,5,6,7}) = per
−1 0 0 0 0
0 −1 0 0 0
1 0 −1 0 0
0 1 0 −1 0
0 0 1 0 −1
= −1.
Weight Choosability of theta Graphs The θ-graphs
The Proof
perA =
6
X
k=1
perASkperA(Sk)
= perA(S1)+ perA(S2)− perA(S5)− perA(S6).
perB =
m1+m2+m3
X
k=1
perB({2},{k})perB({2},{k})
= perB({2},{1})− perB({2},{3}).
Weight Choosability of theta Graphs The θ-graphs
The Proof
perA(S1)+ perA(S2)= perB({2},{1}), perA(S5)+ perA(S6) = perB({2},{3})
perA = perB.
perAθ(m1+4,m2,m3) = perAθ(m1,m2,m3).
Weight Choosability of theta Graphs The θ-graphs
The Table of perA
θ(m1,m2,m3)m1 = 1 m3 = 2 m3 = 3 m3 = 4 m3 = 5 m3 = 6
m2 = 2 0 4 0 4 0
m2 = 3 0 −4 0 4
m2 = 4 0 −4 0
m2 = 5 0 4
m2 = 6 0
Weight Choosability of theta Graphs The θ-graphs
The Table of perA
θ(m1,m2,m3)m1 = 2 m3 = 2 m3 = 3 m3 = 4 m3 = 5 m3 = 6
m2 = 2 −20 0 4 0 −20
m2 = 3 4 0 4 0
m2 = 4 −4 0 4
m2 = 5 4 0
m2 = 6 −20
Weight Choosability of theta Graphs The θ-graphs
The Table of perA
θ(m1,m2,m3)m1 = 3 m3 = 3 m3 = 4 m3 = 5 m3 = 6
m2 = 3 0 −4 0 4
m2 = 4 0 −4 0
m2 = 5 0 4
m2 = 6 0
Weight Choosability of theta Graphs The θ-graphs
The Table of perA
θ(m1,m2,m3)m1 = 4 m3 = 4 m3 = 5 m3 = 6
m2 = 4 20 0 −4
m2 = 5 −4 0
m2 = 6 4
Weight Choosability of theta Graphs The θ-graphs
The Table of perA
θ(m1,m2,m3)m1 = 5 m3 = 5 m3 = 6
m2 = 5 0 4
m2 = 6 0
m1 = 6 m3 = 6 m2 = 6 −20
Weight Choosability of theta Graphs The θ-graphs
2-Choosability
Theorem
Let Aθ(m1,m2,m3) be the associated matrix of θ(m1, m2, m3).
Then perAθ(m1,m2,m3) 6= 0 if and only if m = m1 + m2+ m3 is even.
Weight Choosability of theta Graphs The θ-graphs
Useful Proposition
Proposition
([2]) Let G be a graph whose edge set can be partitioned into two subgraph P, Q, in which P = {e1, e2, ..., em}.
Assume that the associated matrices AP, AQ have permanent indexes at most 2. Let perAP(K ) 6= 0 where K = (k1, k2, ..., km) with ki = 0 for any correspondent edge ei incident to Q. Then pind(AG) ≤ 2.
Weight Choosability of theta Graphs The θ-graphs
The Proof
We can separate P into two parts:
P1 = {ei ∈ P : ei does not link to Q}, P2 = {ei ∈ P : ei link to Q}.
AG =
AP1 ... 0 ... AP2 ...
0 ... AQ
.
Weight Choosability of theta Graphs The θ-graphs
The Proof
By assumption, all the edges ei in P2 gives ki = 0.
AG(K′) =AP(K ) ...
0 AQ(K(Q))
with permanent
perAG(K′) = perAP(K ) × perAQ(K(Q)) 6= 0.
Weight Choosability of theta Graphs The θ-graphs
Consequences
m1 ≤ m2 ≤ m3.
If m3 ≥ 4, then P = Pm3 and Q = Cm1+m2. Check the case m3 ≤ 3.
Weight Choosability of theta Graphs The θ-graphs
3-Choosability
m1 m2 m3 K perAθ(m1,m2,m3)(K )
1 2 2 (0, 2, 1, 1, 1) 12
1 2 3 (1, 1, 1, 1, 1, 1) 4
1 3 3 (2, 0, 1, 1, 1, 1, 1) 16 2 2 2 (1, 1, 1, 1, 1, 1) −20 2 2 3 (2, 1, 0, 1, 1, 1, 1) −4 2 3 3 (1, 1, 1, 1, 1, 1, 1, 1) 4 3 3 3 (2, 2, 1, 0, 0, 1, 1, 1, 1) 4
.
Weight Choosability of theta Graphs The Generalized θ-graphs
What Is a Generalized θ-graph?
Generalized θ-graph θ(m1, m2, ..., mp):
p paths which have the two common endpoints.
In particular, θ(m) = Pm and θ(m1, m2) = Cm1+m2.
Weight Choosability of theta Graphs The Generalized θ-graphs
A Useful Theorem
Theorem
([2]) If G 6= K2 is a clique, complete bipartite graph, or a tree, then mind(G ) ≤ 2.
Weight Choosability of theta Graphs The Generalized θ-graphs
Step 1
Step 1. θ(2, 2, ..., 2) = K2,p is 3-edge weight choosable.
Weight Choosability of theta Graphs The Generalized θ-graphs
Step 2
Step 2. mi ≥ 2 for all i = 1, 2, ...p
θ-graph θ(m1, m2, ..., mp) with is 3-edge weight choosable.
Weight Choosability of theta Graphs The Generalized θ-graphs
Step 3
Step 3. θ(m1, m2, ..., mp, 1) is 3-edge weight choosable.
Weight Choosability of theta Graphs The Generalized θ-graphs
Step 3
Let L ⊆ R and c ∈ R.
L+ c = {l + c : l ∈ L}.
Arbitrary choose x ∈ L(u0u1) and fix this x. p ≥ 3, define a lists L′(e) on θ(m1, m2, ..., mp) by
L′(e) = L(e) + x p− 2.
Weight Choosability of theta Graphs The Generalized θ-graphs
Odd Cycle Absorbs P
3Theorem
Assume that k ≥ 3. Let G = (V , E ) be a graph. Suppose there are path P3 = v0v1v2v3 and odd cycle Ct in G such that P3∩ Ct = {v0, v3} ⊂ V . If G − P3 is k-edge weight choosable, then so is G .
Weight Choosability of theta Graphs The Generalized θ-graphs
The Proof
Weight Choosability of theta Graphs The Generalized θ-graphs
The Proof
Weight Choosability of theta Graphs The Generalized θ-graphs
The Proof
Weight Choosability of theta Graphs The Generalized θ-graphs
The Proof
Weight Choosability of theta Graphs The Generalized θ-graphs
The Proof
Weight Choosability of theta Graphs The Generalized θ-graphs
The Proof
Weight Choosability of theta Graphs The Generalized θ-graphs
The Proof
Weight Choosability of theta Graphs The Generalized θ-graphs
The Proof
Weight Choosability of theta Graphs The Generalized θ-graphs
The Proof
Weight Choosability of theta Graphs The Generalized θ-graphs
The Proof
Weight Choosability of theta Graphs The Generalized θ-graphs
The Proof
L′(e) =
L(e) + xv0v12 +xv2v32 , if e = ui−1ui for odd i ≤ s;
L(e) −xv0v12 − xv2v32 , if e = ui−1ui for even i < s;
L(e) + xv0v12 − xv2v32 , if e = ui−1ui for odd i > s;
L(e) −xv0v12 + xv2v32 , if e = ui−1ui for even i > s;
L(e), otherwise.
Weight Choosability of theta Graphs The Generalized θ-graphs
The Proof
xe =
xe′ − xv0v12 − xv2v32 , if e = ui−1ui for odd i ≤ s;
xe′ + xv0v12 +xv2v32 , if e = ui−1ui for even i < s;
xe′ − xv0v12 +xv2v32 , if e = ui−1ui for odd i > s;
xe′ + xv0v12 −xv2v32 , if e = ui−1ui for even i > s;
xe′, otherwise
Weight Choosability of theta Graphs Further Problems
Problems
1. Total Weight Choosability, by T. Wong and X. Zhu [9].
2. Weight Choosability of Hypergraphs,
by M. Kalkowski, M. Karonski, and F. Pfender [10].
Weight Choosability of theta Graphs References
References
Weight Choosability of theta Graphs References
References
Weight Choosability of theta Graphs References
References
Weight Choosability of theta Graphs References
References
Weight Choosability of theta Graphs References