arXiv Science⌕ Search

arXiv · 0704.0199

Decomposition numbers for finite Coxeter groups and generalised non-crossing partitions

Abstract

Given a finite irreducible Coxeter group $W$, a positive integer $d$, and types $T_1,T_2,...,T_d$ (in the sense of the classification of finite Coxeter groups), we compute the number of decompositions $c=\si_1\si_2 cdots\si_d$ of a Coxeter element $c$ of $W$, such that $\si_i$ is a Coxeter element in a subgroup of type $T_i$ in $W$, $i=1,2,...,d$, and such that the factorisation is "minimal" in the sense that the sum of the ranks of the $T_i$'s, $i=1,2,...,d$, equals the rank of $W$. For the exceptional types, these decomposition numbers have been computed by the first author. The type $A_n$ decomposition numbers have been computed by Goulden and Jackson, albeit using a somewhat different language. We explain how to extract the type $B_n$ decomposition numbers from results of Bóna, Bousquet, Labelle and Leroux on map enumeration. Our formula for the type $D_n$ decomposition numbers is new. These results are then used to determine, for a fixed positive integer $l$ and fixed integers $r_1\le r_2\le ...\le r_l$, the number of multi-chains $π_1\le π_2\le ...\le π_l$ in Armstrong's generalised non-crossing partitions poset, where the poset rank of $π_i$ equals $r_i$, and where the "block structure" of $π_1$ is prescribed. We demonstrate that this result implies all known enumerative results on ordinary and generalised non-crossing partitions via appropriate summations. Surprisingly, this result on multi-chain enumeration is new even for the original non-crossing partitions of Kreweras. Moreover, the result allows one to solve the problem of rank-selected chain enumeration in the type $D_n$ generalised non-crossing partitions poset, which, in turn, leads to a proof of Armstrong's $F=M$ Conjecture in type $D_n$.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Christian Krattenthaler, Thomas Müller. 2009-11-17. Decomposition numbers for finite Coxeter groups and generalised non-crossing partitions. https://arxiv.org/abs/0704.0199

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

Notes on the Divisibility of Divisor Sums of Catalan Numbers

We investigate divisibility questions for the sum-of-divisors function $σ(C_n)$ of the Catalan numbers $C_n=(n+1)^{-1}\binom{2n}{n}$. Using the prime factors occurring to exponent one in $C_n$, we give explicit divisors of $σ(C_n)$ and obtain, for each fixed integer $m\geq 2$, a power $m^{R_m(n)}$ dividing $σ(C_n)$, with $R_m(n)\sim\frac{2\log 2}{φ(m)}\frac{n}{\log n}$. We further prove uniform divisibility for all moduli up to fixed powers of $\log n$ and, with the same leading constant $2\log 2$, for almost all moduli up to $\sqrt{n}$ apart from logarithmic factors. We also establish a complete classification of the integers $m\geq 2$ for which $m\midσ(mk-1)$ for every $k\geq 1$, identify precisely the exceptional coprimality relations among the associated linear forms, and record a twin-prime product divisor of $σ(C_n)$. The arguments use standard prime-distribution results and do not assume any conjecture concerning twin primes.

math.CO↗

Sign-coherence and tropical sign pattern for rank $3$ real cluster-cyclic exchange matrices

The sign-coherence about $c$-vectors was conjectured by Fomin-Zelevinsky and proved by Gross-Hacking-Keel-Kontsevich for integer skew-symmetrizable case. We prove this conjecture associated with $c$-vectors for rank 3 real cluster-cyclic skew-symmetrizable case. Simultaneously, we establish their self-contained recursion and monotonicity. Then, these $c$-vectors are proved to be roots of certain quadratic equations. Based on these results, we prove that the corresponding exchange graphs of $C$-pattern and $G$-pattern are $3$-regular trees. We also study the structure of tropical signs and equip the dihedral group $\mathrm{D}_6$ with a cluster realization via certain mutations.

math.CO↗

Subexponential upper bound on the number of rich words

Let $R_q(n)$ denote the number of rich words of length $n$ over a fixed alphabet of size $q\geq2$. It is known that $R_q(n)^{1/n}\to1$. We establish a nonrecursive subexponential upper bound on the number of rich words, with its asymptotic growth quantified in terms of an iterated logarithm. For every $λ\in(1/2,1)$ and $γ>1$, there exists $c>0$ such that, for every positive function $ϕ$ satisfying $ϕ(n)\to\infty$ and $n/ϕ(n)\to\infty$, the functions \[ f(n)=\left[c\ln^{*}\!\left(\frac{n}{ϕ(n)}\ln q\right)\right]^{1/γ}, \qquad B(n)=q^{\frac{n}{ϕ(n)}+\frac{n}{(2λ)^{f(n)-1}}} \] satisfy $R_q(n)\leq B(n)$ for all sufficiently large $n$ and $B(n)^{1/n}\to1$. Here $\ln^{*}$ is the iterated natural logarithm, with value zero on $(0,1]$. The proof iterates a counting inequality based on palindromic factorization and controls the resulting multiplicative constants using the iterated logarithm.

math.CO↗