partition
6235 papers tagged with this keyword
A Practical Regularity Partitioning Algorithm and its Applications in Clustering
In this paper we introduce a new clustering technique called Regularity Clustering. This new technique is based on the practical variants of the two constructive versions of the Regularity Lemma, a very useful tool in graph theory. The lemma claims that every graph can be partitioned into pseudo-random graphs. While the Regularity Lemma has become very important in proving theoretical results, it has no direct practical applications so far. An important reason for this lack of practical applications is that the graph under consideration has to be astronomically large. This requirement makes its application restrictive in practice where graphs typically are much smaller. In this paper we propose modifications of the constructive versions of the Regularity Lemma that work for smaller graphs as well. We call this the Practical Regularity partitioning algorithm. The partition obtained by this is used to build the reduced graph which can be viewed as a compressed representation of the original graph. Then we apply a pairwise clustering method such as spectral clustering on this reduced graph to get a clustering of the original graph that we call Regularity Clustering. We present results of using Regularity Clustering on a number of benchmark datasets and compare them with standard clustering techniques, such as $k$-means and spectral clustering. These empirical results are very encouraging. Thus in this paper we report an attempt to harness the power of the Regularity Lemma for real-world applications.
A partition of the hypercube into maximally nonparallel Hamming codes
Published in J. Comb. Des. 22(4) 2014, 179-187
• View Publication
• BIB
By using the Gold map, we construct a partition of the hypercube into cosets of Hamming codes such that for every two cosets the corresponding Hamming codes are maximally nonparallel, that is, their intersection cardinality is as small as possible to admit nonintersecting cosets.
Another proof of a Gowers theorem
W. T. Gowers proved that every Lipschitz function from the unit sphere of the Banach space $c_0$ to $\mathbb{R}$ is oscilation stable. His proof uses a result about finite partitions of the set $FIN_k$ of finitely supported functions $p$ from $\mathbb{N}$ to $\{0,1,...,k\}$ with $k$ in $Im(p)$. Every known proof of this fact uses methods of topological dynamics on the space $β\mathbb{N}$ of ultrafilters on $\mathbb{N}$. We give a purely combinatorial proof of this result avoiding the use of ultrafilters.
Threshold functions for distinct parts: revisiting Erdos-Lehner
Published
• View Publication
• BIB
We study four problems: put $n$ distinguishable/non-distinguishable balls into $k$ non-empty distinguishable/non-distinguishable boxes randomly. What is the threshold function $k=k(n) $ to make almost sure that no two boxes contain the same number of balls? The non-distinguishable ball problems are very close to the Erd\H os--Lehner asymptotic formula for the number of partitions of the integer $n$ into $k$ parts with $k=o(n^{1/3})$. The problem is motivated by the statistics of an experiment, where we only can tell whether outcomes are identical or different.
Schubert problems with respect to osculating flags of stable rational curves
Published
• View Publication
• BIB
Given a point z in P^1, let F(z) be the osculating flag to the rational normal curve at point z. The study of Schubert problems with respect to such flags F(z_1), F(z_2), ..., F(z_r) has been studied both classically and recently, especially when the points z_i are real. Since the rational normal curve has an action of PGL_2, it is natural to consider the points (z_1, ..., z_r) as living in the moduli space of r distinct point in P^1 -- the famous M_{0,r}. One can then ask to extend the results on Schubert intersections to the compactification \bar{M}_{0,r}.
The first part of this paper achieves this goal. We construct a flat, Cohen-Macaulay family over \bar{M}_{0,r}, whose fibers over M_{0,r} are isomorphic to G(d,n) and, given partitions lambda_1, ..., lambda_r, we construct a flat Cohen-Macualay family over \bar{M}_{0,r} whose fiber over (z_1, ..., z_r) in M_{0,r} is the intersection of the Schubert varieties indexed by lambda_i with respect to the osculating flags F(z_i).
In the second part of the paper, we investigate the topology of the real points of our family, in the case that sum |lambda_i| = dim G(d,n). We show that our family is a finite covering space of \bar{M}_{0,r}, and give an explicit CW decomposition of this cover whose faces are indexed by objects from the theory of Young tableaux.
Characterizing partition functions of the spin model by rank growth
Published
• View Publication
• BIB
We characterize which graph invariants are partition functions of a spin model over the complex numbers, in terms of the rank growth of associated `connection matrices'.
Gog, Magog and Schützenberger II: Left trapezoids
Published
• View Publication
• BIB
We are interested in finding an explicit bijection between two families of combinatorial objects: Gog and Magog triangles. These two families are particular classes of Gelfand-Tsetlin triangles and are respectively in bijection with alternating sign matrices (ASM) and totally symmetric self complementary plane partitions (TSSCPP). For this purpose, we introduce left Gog and GOGAm trapezoids. We conjecture that these two families of trapezoids are equienumerated and we give an explicit bijection between the trapezoids with one or two diagonals.
Ising Model Observables and Non-Backtracking Walks
Published in J. Math. Phys. 55, 083304 (2014)
• View Publication
• BIB
This paper presents an alternative proof of the connection between the partition function of the Ising model on a finite graph $G$ and the set of non-backtracking walks on $G$. The techniques used also give formulas for spin-spin correlation functions in terms of non-backtracking walks. The main tools used are Viennot's theory of heaps of pieces and turning numbers on surfaces.
Basis problem for analytic multiple gaps
A k-gap is a finite k-sequence of pairwise disjoint monotone families of infinite subsets of N mixed in such a way that we cannot find a partition of N such that each family is trival on one piece of the partition. We prove that, relative to the comparison given by restriction to infinite subsets of N, for every positive integer k there is a finite basis for the class of all analytic k-gaps . We also build the fine structure theory of analytic k-gaps and give some applications.
The content of Chapter 1 of this manuscript have been published as:
A. Avilés, S. Todorcevic, Finite basis for analytic multiple gaps, Publ. Math. IHES. 121 (2015), 57-79.
The content of Chapter 2 (except some technical results from 2.5 and 2.6) and Section 3.1, largely revised and improved, has ben published as:
A. Avilés, S. Todorcevic, Types in the n-adic tree and minimal analytic gaps, Adv. Math. 292 (2016), 558-600.
The content of Sections 3.4, 4.1 and 4.3 have been published as:
A. Avilés, S. Todorcevic, Isolating subgaps of a multiple gap, Monatsh. Math. 186 (2018), 373--392.
The rest of contents may appear elsewhere.
Homology of Littlewood complexes
Published in Selecta Math. (N.S.) 19 (2013), no. 3, 655-698
• View Publication
• BIB
Let V be a symplectic vector space of dimension 2n. Given a partition λwith at most n parts, there is an associated irreducible representation S_{[λ]}(V) of Sp(V). This representation admits a resolution by a natural complex L^λ, which we call the Littlewood complex, whose terms are restrictions of representations of GL(V). When λhas more than n parts, the representation S_{[λ]}(V) is not defined, but the Littlewood complex L^λstill makes sense. The purpose of this paper is to compute its homology. We find that either L^λis acyclic or that it has a unique non-zero homology group, which forms an irreducible representation of Sp(V). The non-zero homology group, if it exists, can be computed by a rule reminiscent of that occurring in the Borel-Weil-Bott theorem. This result can be interpreted as the computation of the "derived specialization" of irreducible representations of Sp(\infty), and as such categorifies earlier results of Koike-Terada on universal character rings. We prove analogous results for orthogonal and general linear groups. Along the way, we will see two topics from commutative algebra: the minimal free resolutions of determinantal ideals and Koszul homology.
Eight-Fifth Approximation for TSP Paths
Published in International Conference on Integer Programming and Combinatorial Optimization IPCO 2013, Lecture Notes in Computer Science, vol 7801. Springer, Berlin, Heidelberg, pp 362-374
• View Publication
• BIB
We prove the approximation ratio 8/5 for the metric $\{s,t\}$-path-TSP problem, and more generally for shortest connected $T$-joins.
The algorithm that achieves this ratio is the simple "Best of Many" version of Christofides' algorithm (1976), suggested by An, Kleinberg and Shmoys (2012), which consists in determining the best Christofides $\{s,t\}$-tour out of those constructed from a family $\Fscr_{>0}$ of trees having a convex combination dominated by an optimal solution $x^*$ of the fractional relaxation. They give the approximation guarantee $\frac{\sqrt{5}+1}{2}$ for such an $\{s,t\}$-tour, which is the first improvement after the 5/3 guarantee of Hoogeveen's Christofides type algorithm (1991). Cheriyan, Friggstad and Gao (2012) extended this result to a 13/8-approximation of shortest connected $T$-joins, for $|T|\ge 4$.
The ratio 8/5 is proved by simplifying and improving the approach of An, Kleinberg and Shmoys that consists in completing $x^*/2$ in order to dominate the cost of "parity correction" for spanning trees. We partition the edge-set of each spanning tree in $\Fscr_{>0}$ into an $\{s,t\}$-path (or more generally, into a $T$-join) and its complement, which induces a decomposition of $x^*$. This decomposition can be refined and then efficiently used to complete $x^*/2$ without using linear programming or particular properties of $T$, but by adding to each cut deficient for $x^*/2$ an individually tailored explicitly given vector, inherent in $x^*$.
A simple example shows that the Best of Many Christofides algorithm may not find a shorter $\{s,t\}$-tour than 3/2 times the incidentally common optima of the problem and of its fractional relaxation.
A Bijection for Crossings and Nestings
For a subclass of matchings, set partitions, and permutations, we describe a direct bijection involving only arc annotated diagrams that not only interchanges maximum nesting and crossing numbers, but also all refinements of crossing and nesting numbers. Furthermore, we show that the bijection cannot be applied to a similar class of coloured arc annotated diagrams.
Symmetric and nonsymmetric Hall-Littlewood polynomials of type BC
Published in J. Algebr. Comb. 42 (2015), 331-364
• Search Publication
Koornwinder polynomials are a 6-parameter BC_n-symmetric family of Laurent polynomials indexed by partitions, from which Macdonald polynomials can be recovered in suitable limits of the parameters. As in the Macdonald polynomial case, standard constructions via difference operators do not allow one to directly control these polynomials at q=0. In the first part of this paper, we provide an explicit construction for these polynomials in this limit, using the defining properties of Koornwinder polynomials. Our formula is a first step in developing the analogy between Hall-Littlewood polynomials and Koornwinder polynomials at q=0. Next, we provide an analogous construction for the nonsymmetric Koornwinder polynomials in the same limiting case. The method employed in this paper is a BC-type adaptation of techniques used in an earlier work of the author, which gave a combinatorial method for proving vanishing results of Rains and Vazirani at the Hall-Littlewood level. As a consequence of this work, we obtain direct arguments for the constant term evaluations and norms in both the symmetric and nonsymmetric cases.
Kronecker coefficients for one hook shape
We give a positive combinatorial formula for the Kronecker coefficient g_{lambda mu(d) nu} for any partitions lambda, nu of n and hook shape mu(d) := (n-d,1^d). Our main tool is Haiman's \emph{mixed insertion}. This is a generalization of Schensted insertion to \emph{colored words}, words in the alphabet of barred letters \bar{1},\bar{2},... and unbarred letters 1,2,.... We define the set of \emph{colored Yamanouchi tableaux of content lambda and total color d} (CYT_{lambda, d}) to be the set of mixed insertion tableaux of colored words w with exactly d barred letters and such that w^{blft} is a Yamanouchi word of content lambda, where w^{blft} is the ordinary word formed from w by shuffling its barred letters to the left and then removing their bars. We prove that g_{lambda mu(d) nu} is equal to the number of CYT_{lambda, d} of shape nu with unbarred southwest corner.
Enumeration of Cylindric Plane Partitions - Part II
Published
• View Publication
• BIB
Cylindric plane partitions may be thought of as a natural generalization of reverse plane partitions. A generating series for the enumeration of cylindric plane partitions was recently given by Borodin. As in the reverse plane partition case, the right hand side of this identity admits a simple factorization form in terms of the "hook lengths" of the individual boxes in the underlying shape. The main result of this paper is a new bijective proof of Borodin's identity which makes use of Fomin's growth diagram framework for generalized RSK correspondences.
Small Littlewood-Richardson coefficients
Published
• View Publication
• BIB
We develop structural insights into the Littlewood-Richardson graph, whose number of vertices equals the Littlewood-Richardson coefficient c(λ,μ,ν) for given partitions λ, μ, and ν. This graph was first introduced by Bürgisser and Ikenmeyer in arXiv:1204.2484, where its connectedness was proved.
Our insights are useful for the design of algorithms for computing the Littlewood-Richardson coefficient: We design an algorithm for the exact computation of c(λ,μ,ν) with running time O(c(λ,μ,ν)^2 poly(n)), where λ, μ, and ν are partitions of length at most n. Moreover, we introduce an algorithm for deciding whether c(λ,μ,ν) >= t whose running time is O(t^2 poly(n)). Even the existence of a polynomial-time algorithm for deciding whether c(λ,μ,ν) >= 2 is a nontrivial new result on its own.
Our insights also lead to the proof of a conjecture by King, Tollu, and Toumazet posed in 2004, stating that c(λ,μ,ν) = 2 implies c(Mλ,Mμ,Mν) = M + 1 for all M. Here, the stretching of partitions is defined componentwise.
How to compute the multigraded Hilbert depth of a module
Published
• View Publication
• BIB
The aim of this paper is to introduce a method for computing Hilbert decompositions (and consequently the Hilbert depth) of a finitely generated multigraded module $M$ over the polynomial ring $K[X_1,..., X_n]$ by reducing the problem to the computation of the finite set of the new defined Hilbert partitions. Moreover, in the last section, we show that Hilbert partitions may also be used for computing the Stanley depth of the module $M$.
$p^\ell$-Torsion Points In Finite Abelian Groups And Combinatorial Identities
Published
• View Publication
• BIB
The main aim of this article is to compute all the moments of the number of $p^\ell$-torsion elements in some type of nite abelian groups. The averages involved in these moments are those de ned for the Cohen-Lenstra heuristics for class groups and their adaptation for Tate-Shafarevich groups. In particular, we prove that the heuristic model for Tate-Shafarevich groups is compatible with the recent conjecture of Poonen and Rains about the moments of the orders of $p$-Selmer groups of elliptic curves. For our purpose, we are led to de ne certain polynomials indexed by integer partitions and to study them in a combinatorial way. Moreover, from our probabilistic model, we derive combinatorial identities, some of which appearing to be new, the others being related to the theory of symmetric functions. In some sense, our method therefore gives for these identities a somehow natural algebraic context.
A Euclid style algorithm for MacMahon's partition analysis
Published in Journal of Combinatorial Theory Series A, 131 (2015), 32-60
• View Publication
• BIB
Solutions to a linear Diophantine system, or lattice points in a rational convex polytope, are important concepts in algebraic combinatorics and computational geometry. The enumeration problem is fundamental and has been well studied, because it has many applications in various fields of mathematics. In algebraic combinatorics, MacMahon's partition analysis has become a general approach for linear Diophantine system related problems. Many algorithms have been developed, but "bottlenecks" always arise when dealing with complex problems. While in computational geometry, Barvinok's important result asserts the existence of a polynomial time algorithm when the dimension is fixed. However, the implementation by the LattE package of De Loera et. al. does not perform well in many situations. By combining excellent ideas in the two fields, we generalize Barvinok's result by giving a polynomial time algorithm for MacMahon's partition analysis in a suitable condition. We also present an elementary Euclid style algorithm, which might not be polynomial but is easy to implement and performs well. As applications, we contribute the generating series for magic squares of order 6.
Open and solved problems concerning polarized partition relations
Published in Fundamenta Mathematicae, vol. 234 no.1, 2016
• View Publication
• BIB
We list some open problems, concerning the polarized partition relation. We solve a couple of them by showing that for every singular cardinal $μ$ one can force the strong polarized relation with respect to the pair $μ^+,μ$.