a new formula for stanley s chromatic symmetric function
play

A new formula for Stanleys chromatic symmetric function for unit - PowerPoint PPT Presentation

A new formula for Stanleys chromatic symmetric function for unit interval graphs and e -positivity for triangular ladder graphs Samantha Dahlberg Arizona State University FPSAC July 2, 2019 Overview 1 Chromatic symmetric functions. 2 The (3


  1. A new formula for Stanley’s chromatic symmetric function for unit interval graphs and e -positivity for triangular ladder graphs Samantha Dahlberg Arizona State University FPSAC July 2, 2019

  2. Overview 1 Chromatic symmetric functions. 2 The (3 + 1)-free conjecture. 3 Triangular ladders, TL n . 4 Chromatic symmetric functions in non-commuting variables. 5 Deletion-contraction. 6 Semi-symmetrized e -positivity. 7 Signed formula for unit interval graphs. 8 Sign-reversing involution for TL n . 9 Further work.

  3. Graphs colorings Given G with vertex set V a proper coloring κ of G is κ : V → { 1 , 2 , 3 , . . . } so if u , v ∈ V are joined by an edge then κ ( u ) � = κ ( v ) . 1 2 ✘ 1 2 ✔

  4. Chromatic symmetric function: Stanley 1995 Given a proper coloring κ of vertices v 1 , . . . , v N we associate a monomial in commuting variables x 1 , x 2 , x 3 , . . . x κ ( v 1 ) x κ ( v 2 ) · · · x κ ( v N ) . 1 2 3 1 2 3 x 2 1 x 2 x 1 x 2 x 3 The chromatic symmetric function is � X G = x κ ( v 1 ) x κ ( v 2 ) · · · x κ ( v N ) κ summed over all proper colorings κ . X P 3 = x 2 1 x 2 + x 1 x 2 2 + x 2 1 x 3 + x 1 x 2 3 + x 2 2 x 3 + x 2 x 2 3 + · · · + 6 x 1 x 2 x 3 + · · ·

  5. Symmetric functions The algebra of symmetric functions , Λ, contains all formal power series f in commuting variables x 1 , x 2 , . . . such that for all permutations π f ( x 1 , x 2 , . . . ) = f ( x π (1) , x π (2) , . . . ) . f ( x ) = x 2 1 x 2 + x 1 x 2 2 + x 2 1 x 3 + x 1 x 2 3 + x 2 2 x 3 + x 2 x 2 3 + · · · Fact: Any basis of Λ is indexed by integer partitions. An integer partition λ = ( λ 1 , . . . , λ ℓ ) of n , λ ⊢ n , is a list of positive integers whose parts λ i weakly decrease and sum to n . (3 , 1 , 1) = (3 , 1 2 ) ⊢ 5

  6. Classic Bases: elementary The i-th elementary symmetric function , e i , is � e i = x j 1 . . . x j i j 1 < j 2 < ··· < j i and e λ = e λ 1 . . . e λ ℓ . e (2 , 1) = e 2 e 1 = ( x 1 x 2 + x 1 x 3 + x 2 x 3 + · · · )( x 1 + x 2 + x 3 + · · · ) For the complete graph K n on n vertices X K n = n ! e n .

  7. e -positivity Call a graph G e-positive if X G is a non-negative sum of elementary symmetric functions. G = has X G = e (2 , 1) + 3 e (3) . ✔ K 31 = has X K 31 = e (2 , 1 , 1) − 2 e (2 , 2) + 5 e (3 , 1) + 4 e (4) . ✘ The claw, K 31 , is the smallest graph that is not e -positive.

  8. (3 + 1) -free conjecture Conjecture (Stanley-Stembridge 1993) If G is an indifference graph of a (3 + 1) -free poset then X G is e-positive. Hasse diagram Indifference graph Theorem (Guay-Paquet 2013) It is sufficient to prove the Stanley-Stembridge conjecture for all (2 + 2) and (3 + 1) -free posets.

  9. Interval graphs The indifference graphs for (2 + 2) and (3 + 1)-free posets are unit interval graphs . Construct an unit interval graphs from a collection of integer intervals [ a 1 , b 1 ] , [ a 2 , b 2 ] , . . . , [ a l , b l ] . On each interval we place a complete graph. The interval graph for the intervals [1 , 5], [4 , 7] and [7 , 8] is 3 4 6 7 8 1 5 2

  10. Known e -positive unit interval graphs The paths [1 , 2] , [2 , 3] , . . . , [ n − 1 , n ] (Stanley 1995). Any list containing [1 , j ] and [ j + 1 , n ] (Stanley 1995). 2 7 5 1 6 9 3 4 8 Any list containing [2 , n − 1] (Cho and Huh 2017). 3 6 7 1 2 4 5 Any list [1 , j 1 ] , [ j 1 , j 2 ] , . . . , [ j k , n ] (Gebhard and Sagan 2001).

  11. Triangular Ladders The graph P n , 2 comes from intervals [1 , 3] , [2 , 4] , . . . , [ n − 2 , n ], which we will call triangular ladders , TL n . The graph TL 8 comes from [1 , 3] , [2 , 4] , [3 , 5] , [4 , 6] , [5 , 7] , [6 , 8] . 1 3 5 7 2 4 6 8 In 1995 Stanley wrote “It remains open whether P d , 2 is e -positive.”

  12. Chromatic symmetric functions in non-commuting variables A generalization by Gebhard and Sagan (2001). Fix an ordering on the vertices vertices v 1 , . . . , v N . The chromatic symmetric function in non-commuting variables is � Y G = x κ ( v 1 ) x κ ( v 2 ) · · · x κ ( v N ) κ summed over all proper colorings κ . G = 1 2 3 Y G = x 1 x 2 x 1 + x 2 x 1 x 2 + x 1 x 3 x 1 + x 3 x 1 x 3 + · · · + x 1 x 2 x 3 + x 1 x 3 x 2 + x 2 x 1 x 3 + x 2 x 3 x 1 + · · · Fact: The vertex labeling matters.

  13. Symmetric functions in non-commuting variables The algebra of symmetric functions in non-commuting variables , NCSym, contains all formal power series f in non-commuting variables x 1 , x 2 , . . . such that for all permutations π f ( x 1 , x 2 , . . . ) = f ( x π (1) , x π (2) , . . . ) . f ( x ) = x 1 x 1 x 2 + x 2 x 2 x 1 + x 1 x 1 x 3 + x 3 x 3 x 1 + x 2 x 2 x 3 + x 3 x 3 x 2 + · · · Fact: Any basis of NCSym is indexed by set partitions. An set partition π = B 1 / B 2 / · · · / B k of [ n ] = { 1 , 2 , . . . n } , π ⊢ [ n ] is a collection of nonempty disjoint subsets B i called blocks that union to [ n ]. { 1 , 4 } / { 2 , 5 } / { 3 } = 14 / 25 / 3 ⊢ [5]

  14. Classic Bases: elementary For π ⊢ [ n ] the elementary symmetric function in non-commuting variables , e π , is � e π = x i 1 x i 2 · · · x i n ( i 1 , i 2 ,..., i n ) where i j � = i k if j and k are in the same block of π . e 12 / 3 = x 1 x 2 x 2 + x 1 x 2 x 1 + · · · + x 1 x 2 x 3 + · · · For the complete graph K n on n vertices Y K n = e 12 ··· n .

  15. Deletion-contraction for Y G To delete an edge ǫ , G − ǫ , remove ǫ . 1 1 ǫ 3 4 3 4 delete ǫ 2 2 To contract an edge ǫ between u and v , G /ǫ , merge u and v and any multiedges created. 1 ǫ 3 4 1 3 4 contract ǫ 2

  16. Deletion-Contraction for Y G ǫ 2 ↑ 2 = 3 − 1 2 3 1 2 1 Y P 3 = ( x 1 x 2 x 1 + x 1 x 2 x 2 + x 1 x 2 x 3 + · · · ) − ( x 1 x 2 x 2 + · · · ) Given a monomial of degree n − 1 define the induced monomial for j < n to be x i 1 x i 2 · · · x i j · · · x i n − 1 ↑ j = x i 1 x i 2 · · · x i j · · · x i n − 1 x i j . Theorem (Gebhard and Sagan 2001) For G with vertices V = [ n ] and an edge ǫ between vertices j and n we have Y G = Y G − ǫ − Y G /ǫ ↑ j .

  17. Induction on monomials Theorem (D 2018) G is a unit interval graph with intervals [ a 1 , 1] , [ a 2 , 2] , . . . , [ a n , n ] and G ′ is G after removing vertex n. Then, n − 1 � Y G ′ ↑ i . Y G = Y G ′ Y K 1 − i = a n 4 = 4 − 3 ↑ 3 − 3 3 ↑ 3 1 2 1 2 1 2 4 − 3 ↑ 2 − 3 ↑ 3 3 = 1 2 1 2 1 2

  18. Semi-symmetrizing e 12 ↑ 1 = 1 ≡ 1 � � � � e 12 / 3 + e 1 / 23 − e 13 / 2 − e 123 e 12 / 3 − e 123 2 2 For π ⊢ [ n ] let λ ( π ) ⊢ n be formed by all the block sizes. λ (1 / 23) = λ (13 / 2) = (2 , 1) and 1 / 23 ∼ 13 / 2 Say two set partitions π ⊢ [ n ] and σ ⊢ [ n ] are related , π ∼ σ , if 1 λ ( π ) = λ ( σ ) and 2 the sizes of the blocks containing n are the same. If π ∼ σ we say e π and e σ are equivalent , e π ≡ e σ . Extend this definition linearly.

  19. Semi-symmetrizing For π ⊢ [ n − 1] define π ⊕ j n ⊢ [ n ] to be the integer partition where we place n in the same block as j . Theorem (Gebhard and Sagan 2001) For π ⊢ [ n − 1] , j < n and b the size of the block in π containing n − 1 we have e π ↑ j ≡ 1 � � e π/ n − e π ⊕ j n . b Call G semi-symmetrized e-positive if Y G ≡ f for some f ∈ NCSym that is a sum of nonnegative e π . Fact: If G is semi-symmetrized e -positive then G is e -positive.

  20. New formula for unit interval graphs Theorem (D 2018) For a unit interval graph G on n vertices, Y G ≡ 1 � ( − 1) t ( D ) e π ( D ) . n ! D ∈A ′ L ( G ) Arc diagrams D ∈ A ′ L ( G ) are defined by: all vertices have at most one left arc, each arc possibly has a tic mark, a permutation labeling increasing on all pieces and one vertex in each right-most piece is marked with a star. D ∈ A ′ L ( TL 9 ) with t ( D ) = 3 and π ( D ) = 13 / 2 / 4 / 57 / 6 / 89. / / ⋆ ⋆ ⋆ / 1 2 3 4 5 6 7 8 9 8 7 9 6 4 3 5 1 2

  21. The sign-reversing involution for TL n The general inductive idea: ϕ ϕ ( D ) D ⋆ ⋆ / / ϕ / / / / ⋆ ⋆ ⋆ ⋆ ⋆ / / / 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 7 6 6 7 8 9 4 3 5 1 2 8 9 4 3 5 1 2 π ( D ) = 13 / 2 / 4 / 57 / 6 / 89 π ( ϕ ( D )) = 1 / 2 / 34 / 57 / 6 / 89 t ( D ) = 3 t ( ϕ ( D )) = 4 The sign-reversing involution: changes t ( D ) by one and has π ( D ) ∼ π ( ϕ ( D )). There are 18 cases where D is a fixed point.

  22. The sign-reversing involution D ∈ A L ( TL 9 ) with π ( D ) = 12346 / 579 / 8 is a fixed point. ⋆ ⋆ ⋆ 1 2 3 4 5 6 7 8 9 5 6 7 8 1 9 2 4 3 Fixed points: have no tic marks, have a star on each connected component and satisfy 5 other more detailed conditions.

  23. New family of e -positive graphs Theorem (D 2018) The triangular ladder TL n , n ≥ 1 , is semi-symmetrized e-positive and so e-positive. Given a graph G on [ n ] and H on [ m ] their concatenation is the graph G · H on [ n + m − 1] where G is on the first n vertices and H is on the last m . 3 6 7 K 4 · TL 4 = 1 2 4 5 Theorem (Gebhard and Sagan 2001) If a graph G is semi-symmetrized e-positive then so is the concatenation G · K m and G · TL 4 .

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend


More recommend