The log-behavior of $$\root n \of {p(n)}$$ and $$\root n \of {p(n)/n}$$p(n)/nn

Let p(n) denote the partition function and let \(\Delta \) be the difference operator with respect to n. In this paper, we obtain a lower bound for \(\Delta ^2\log \root n-1 \of {p(n-1)/(n-1)}\), leading to a proof of a conjecture of Sun on the log-convexity of \(\{\root n \of {p(n)/n}\}_{n\ge 60}\). Using the same argument, it can be shown that for any real number \(\alpha \), there exists an integer \(n(\alpha )\) such that the sequence \(\{\root n \of {p(n)/n^{\alpha }}\}_{n\ge n(\alpha )}\) is log-convex. Moreover, we show that \(\lim \limits _{n \rightarrow +\infty }n^{\frac{5}{2}}\Delta ^2\log \root n \of {p(n)}=3\pi /\sqrt{24}\). Finally, by finding an upper bound for \(\Delta ^2 \log \root n-1 \of {p(n-1)}\), we establish an inequality on the ratio \(\frac{\root n-1 \of {p(n-1)}}{\root n \of {p(n)}}\).     

Une formule asymptotique pour ?n \leqq x az (n)d (n + 1)\sum\limits_{n \leqq x} {a_z (n)d} (n + 1)

Sans résumé     

Processus de diffusion associe aux mesures de Gibbs sur \mathbb{Z}^d

Sans résumé     

On the rational recursive sequencex_{n + 1} = \frac{{\alpha x_n + \beta x_{n - 1} + \gamma x_{n - 2} + \delta x_{n - 3} }}{{Ax_n + Bx_{n - 1} + Cx_{n - 2} + Dx_{n - 3} }}

The main objective of this paper is to study the boundedness character, the periodic character and the global stability of the positive solutions of the following difference equation $x_{n + 1} = \frac{{\alpha x_n + \beta x_{n - 1} + \gamma x_{n - 2} + \delta x_{n - 3} }}{{Ax_n + Bx_{n - 1} + Cx_{n - 2} + Dx_{n - 3} }},n = 0,1,2.....$ where the coefficientsA, B, C, D, α, β, γ, δ, and the initial conditionsx _-3,x _-2,x _-1,x ₀ are arbitrary positive real numbers.     

Improved Approximation Algorithms for MAX \frac{n}{{\text{2}}}-DIRECTED-BISECTION and MAX \frac{n}{{\text{2}}}-DENSE-SUBGRAPH

We consider the MAX $\frac{n}{{\text{2}}}$ -DIRECTED-BISECTION problem, i.e., partitioning the vertices of a directed graph into two blocks of equal cardinality so as to maximize the total weight of the edges in the directed cut. A polynomial approximation algorithm using a semidefinite relaxation with 0.6458 performance guarantee is presented for the problem. The previous best-known results for approximating this problem are 0.5 using a linear programming relaxation, 0.6440 using a semidefinite relaxation. We also consider the MAX $\frac{n}{{\text{2}}}$ -DENSE-SUBGRAPH problem, i.e., determine a block of half the number of vertices from a weighted undirected graph such that the sum of the edge weights, within the subgraph induced by the block, is maximized. We present an 0.6236 approximation of the problem as opposed to 0.6221 of Halperin and Zwick.     

Über die Reihe\sum\limits_{n = 0}^\infty {q^{n^2 } x^n \cdot }

Ohne ZusammenfassungAus einem an Herrn G. Pólya in Zürich gerichteten Briefe.     

The rigidity of Clifford torusS^1 \left( {\sqrt {\frac{1}{n}} } \right) \times S^{n - 1} \left( {\sqrt {\frac{{n - 1}}{n}} } \right)

In this paper, we prove that ifM is ann-dimensional closed minimal hypersurface with two distinct principal curvatures of a unit sphereS ⁿ⁺¹ (1), thenS=n andM is a Clifford torus ifn≤S≤n+[2n ²(n+4)/3(n(n+4)+4)], whereS is the squared norm of the second fundamental form ofM.     

Sur la régularitéC ∞ du\bar \partial au bord d'un domaine de ℂ n dont la forme de Levi a exactements valeurs propres strictement négatives

Sans résumé

---

     

Randomized Operators on {n\times n} Matrices and Applications

Some combinatorial and probabilistic estimates motivated by earlier works due to S. Kwapien and C. Schütt are proved. We study these estimates in the general setting of rearrangement invariant function and sequence spaces and identify the class of function spaces in which such estimates hold. We demonstrate the sharpness of our results and present some applications, one of which is an alternative proof of a familiar Raynaud–Schütt theorem describing symmetric subspaces in \({L_1}\).     

Asymptotic behavior of multiperiodic functionsG(x) = \prod\limits_{n = 1}^\infty {g(x/2^n )}

Let 0≤g be a dyadic Hölder continuous function with period 1 and g(0)=1, and let $G(x) = \prod\nolimits_{n = 0}^\infty {g(x/{\text{2}}^n )} $ . In this article we investigate the asymptotic behavior of $\smallint _0^{\rm T} \left| {G(x)} \right|^q dx$ and $\frac{1}{n}\sum\nolimits_{k = 0}^n {\log g(2^k x)} $ using the dynamical system techniques: the pressure function and the variational principle. An algorithm to calculate the pressure is presented. The results are applied to study the regulatiry of wavelets and Bernoulli convolutions.     

On the Diophantine equation \dfrac{ax^{n+2l}+c}{abt^{2}x^{n}+c}{\displaystyle\frac{ax^{n+2l}+c}{abt^{2}x^{n}+c}}=by^{2}

We obtain all solutions of the equation $\frac{ax^{n+2l}+c}{abt^{2}x^{n}+c} = by^{2}$ with c??{??1,??2,??4}.     

A New Proof of Diophantine Equation $\Bigg( \begin{matrix} n \\ 2 \end{matrix} \Bigg) = \Bigg( \begin{matrix} m \\ 4 \end{matrix} \Bigg)$

By using algebraic number theory and $p$-adic analysis method, we give a new and simple proof of Diophantine equation $\Bigg( \begin{matrix} n \\ 2 \end{matrix} \Bigg) =\Bigg( \begin{matrix} m \\ 4 \end{matrix} \Bigg)$.     

Biharmonic Submanifolds with Parallel Mean Curvature in \mathbb{S}^{n}\times\mathbb{R}

We find a Simons type formula for submanifolds with parallel mean curvature vector (pmc submanifolds) in product spaces M ⁿ (c)×?, where M ⁿ (c) is a space form with constant sectional curvature c, and then we use it to prove a gap theorem for the mean curvature of certain complete proper-biharmonic pmc submanifolds, and classify proper-biharmonic pmc surfaces in $\mathbb{S}^{n}(c)\times\mathbb{R}$ .     

Commuting involution graphs for \tilde{A}_{n}

In this article we consider the commuting graphs of involution conjugacy classes in the affine Weyl group We show that where the graph is connected the diameter is at most 6. Received: 24 February 2005     

Combinatorics of character formulas for the lie superalgebra \mathfrak{g}\mathfrak{l}\left( {m,n} \right)

Let \mathfrakg \mathfrak{g} be the Lie superalgebra \mathfrakg\mathfrakl( m,n ) \mathfrak{g}\mathfrak{l}\left( {m,n} \right) . Algorithms for computing the composition factors and multiplicities of Kac modules for \mathfrakg \mathfrak{g} were given by the second author, [12] and by J. Brundan [1]. We give a combinatorial proof of the equivalence between the two algorithms. The proof uses weight and cap diagrams introduced by Brundan and C. Stroppel, and cancelations between paths in a graph G \mathcal{G} defined using these diagrams. Each vertex of G \mathcal{G} corresponds to a highest weight of a finite dimensional simple module, and each edge is weighted by a nonnegative integer. If E \mathcal{E} is the subgraph of G \mathcal{G} obtained by deleting all edges of positive weight, then E \mathcal{E} is the graph that describes nonsplit extensions between simple highest weight modules. We also give a procedure for finding the composition factors of any Kac module, without cancelation. This procedure leads to a second proof of the main result.     

Brauer algebras of type \mathrm{I}_2^n(n\ge 5)

We will present an algebra related to the Coxeter group of type \(\mathrm{I}_2^n\) which can be taken as the twisted subalgebra in Brauer algebra of type \(\mathrm{A}_{n-1}\) . Also, we will describe some properties of this algebra.     

A note on the congruence \left( {_{mp^k }^{np^k } } \right) \equiv \left( {_m^n } \right) (mod p r )

In the paper we discuss the following type congruences: $$\left( {_{mp^k }^{np^k } } \right) \equiv \left( {_m^n } \right)(\bmod p^r ),$$ where p is a prime, n, m, k and r are various positive integers with n ? m ? 1, k ? 1 and r ? 1. Given positive integers k and r, denote by W(k, r) the set of all primes p such that the above congruence holds for every pair of integers n ? m ? 1. Using Ljunggren’s and Jacobsthal’s type congruences, we establish several characterizations of sets W(k, r) and inclusion relations between them for various values k and r. In particular, we prove that W(k + i, r) = W(k ? 1, r) for all k ? 2, i ? 0 and 3 ? r ? 3k, and W(k, r) = W(1, r) for all 3 ? r ? 6 and k ? 2. We also noticed that some of these properties may be used for computational purposes related to congruences given above.