A counterexample concerning the -projector onto linear spline spaces

For the -orthogonal projection onto spaces of linear splines over simplicial partitions in polyhedral domains in , , we show that in contrast to the one-dimensional case, where independently of the nature of the partition, in higher dimensions the -norm of cannot be bounded uniformly with respect to the partition. This fact is folklore among specialists in finite element methods and approximation theory but seemingly has never been formally proved.

     

Uniform error estimates in the finite element method for a singularly perturbed reaction-diffusion problem

Consider the problem with homogeneous Neumann boundary condition in a bounded smooth domain in . The whole range is treated. The Galerkin finite element method is used on a globally quasi-uniform mesh of size ; the mesh is fixed and independent of .

A precise analysis of how the error at each point depends on and is presented. As an application, first order error estimates in , which are uniform with respect to , are given.

     

On Meinardus' examples for the conjugate gradient method

The conjugate gradient (CG) method is widely used to solve a positive definite linear system of order . It is well known that the relative residual of the th approximate solution by CG (with the initial approximation ) is bounded above by

   with

where is 's spectral condition number. In 1963, Meinardus (Numer. Math., 5 (1963), pp. 14-23) gave an example to achieve this bound for but without saying anything about all other . This very example can be used to show that the bound is sharp for any given by constructing examples to attain the bound, but such examples depend on and for them the th residual is exactly zero. Therefore it would be interesting to know if there is any example on which the CG relative residuals are comparable to the bound for all . There are two contributions in this paper:

1. A closed formula for the CG residuals for all on Meinardus' example is obtained, and in particular it implies that the bound is always within a factor of of the actual residuals;
2. A complete characterization of extreme positive linear systems for which the th CG residual achieves the bound is also presented.

     

Explicit values of multi-dimensional Kloosterman sums for prime powers, II

For any integer fix , and let denote the group of reduced residues modulo . Let , a power of a prime . The hyper-Kloosterman sums of dimension are defined for by

where denotes the multiplicative inverse of modulo .

Salie evaluated in the classical setting for even , and for odd with . Later, Smith provided formulas that simplified the computation of in these cases for . Recently, Cochrane, Liu and Zheng computed upper bounds for in the general case , stopping short of their explicit evaluation. Here I complete the computation they initiated to obtain explicit values for the Kloosterman sums for , relying on basic properties of some simple specialized exponential sums. The treatment here is more elementary than the author's previous determination of these Kloosterman sums using character theory and -adic methods. At the least, it provides an alternative, independent evaluation of the Kloosterman sums.

     

On the embedding problem for representations

Let denote the double cover of corresponding to the element in where transpositions lift to elements of order and the product of two disjoint transpositions to elements of order . Given an elliptic curve , let denote its -torsion points. Under some conditions on elements in correspond to Galois extensions of with Galois group (isomorphic to) . In this work we give an interpretation of the addition law on such fields, and prove that the obstruction for having a Galois extension with gives a homomorphism . As a corollary we can prove (if has conductor divisible by few primes and high rank) the existence of -dimensional representations of the absolute Galois group of attached to and use them in some examples to construct modular forms mapping via the Shimura map to (the modular form of weight attached to) .

     

A search for Fibonacci-Wieferich and Wolstenholme primes

A prime is called a Fibonacci-Wieferich prime if , where is the th Fibonacci number. We report that there exist no such primes . A prime is called a Wolstenholme prime if . To date the only known Wolstenholme primes are 16843 and 2124679. We report that there exist no new Wolstenholme primes . Wolstenholme, in 1862, proved that for all primes . It is estimated by a heuristic argument that the ``probability' that is Fibonacci-Wieferich (independently: that is Wolstenholme) is about . We provide some statistical data relevant to occurrences of small values of the Fibonacci-Wieferich quotient modulo .

     

Efficient CM-constructions of elliptic curves over finite fields

We present an algorithm that, on input of an integer together with its prime factorization, constructs a finite field and an elliptic curve over for which has order . Although it is unproved that this can be done for all , a heuristic analysis shows that the algorithm has an expected run time that is polynomial in , where is the number of distinct prime factors of . In the cryptographically relevant case where is prime, an expected run time can be achieved. We illustrate the efficiency of the algorithm by constructing elliptic curves with point groups of order and nextprime.

     

The Lucas-Pratt primality tree

In 1876, E. Lucas showed that a quick proof of primality for a prime could be attained through the prime factorization of and a primitive root for . V. Pratt's proof that PRIMES is in NP, done via Lucas's theorem, showed that a certificate of primality for a prime could be obtained in modular multiplications with integers at most . We show that for all constants , the number of modular multiplications necessary to obtain this certificate is greater than for a set of primes with relative asymptotic density 1.

     

Modified Taylor reproducing formulas and h-p clouds

We study two different approximations of a multivariate function by operators of the form , where is an -reproducing partition of unity and are modified Taylor polynomials of degree expanded at . The first approximation was introduced by Xuli (2003) in the univariate case and generalized for convex domains by Guessab et al. (2005). The second one was introduced by Duarte (1995) and proved in the univariate case. In this paper, we first relax the Guessab's convexity assumption and we prove Duarte's reproduction formula in the multivariate case. Then, we introduce two related reproducing quasi-interpolation operators in Sobolev spaces. A weighted error estimate and Jackson's type inequalities for h-p cloud function spaces are obtained. Last, numerical examples are analyzed to show the approximative power of the method.

     

New techniques for bounds on the total number of prime factors of an odd perfect number

Let denote the sum of the positive divisors of . We say that is perfect if . Currently there are no known odd perfect numbers. It is known that if an odd perfect number exists, then it must be of the form , where are distinct primes and . Define the total number of prime factors of as . Sayers showed that . This was later extended by Iannucci and Sorli to show that . This was extended by the author to show that . Using an idea of Carl Pomerance this paper extends these results. The current new bound is .

     

A dual finite element complex on the barycentric refinement

Given a two dimensional oriented surface equipped with a simplicial mesh, the standard lowest order finite element spaces provide a complex centered on Raviart-Thomas divergence conforming vector fields. It can be seen as a realization of the simplicial cochain complex. We construct a new complex of finite element spaces on the barycentric refinement of the mesh which can be seen as a realization of the simplicial chain complex on the original (unrefined) mesh, such that the duality is non-degenerate on for each . In particular is a space of -conforming vector fields which is dual to Raviart-Thomas -conforming elements. When interpreted in terms of differential forms, these two complexes provide a finite-dimensional analogue of Hodge duality.

     

Asymptotically fast group operations on Jacobians of general curves

Let be a curve of genus over a field . We describe probabilistic algorithms for addition and inversion of the classes of rational divisors in the Jacobian of . After a precomputation, which is done only once for the curve , the algorithms use only linear algebra in vector spaces of dimension at most , and so take field operations in , using Gaussian elimination. Using fast algorithms for the linear algebra, one can improve this time to . This represents a significant improvement over the previous record of field operations (also after a precomputation) for general curves of genus .

     

On univoque Pisot numbers

We study Pisot numbers which are univoque, i.e., such that there exists only one representation of as , with . We prove in particular that there exists a smallest univoque Pisot number, which has degree . Furthermore we give the smallest limit point of the set of univoque Pisot numbers.

     

Divisors in residue classes, constructively

Let be integers satisfying , , , and let . Lenstra showed that the number of integer divisors of equivalent to is upper bounded by . We re-examine this problem, showing how to explicitly construct all such divisors, and incidentally improve this bound to .

     

Parity-regular Steinhaus graphs

Steinhaus graphs on vertices are certain simple graphs in bijective correspondence with binary -sequences of length . A conjecture of Dymacek in 1979 states that the only nontrivial regular Steinhaus graphs are those corresponding to the periodic binary sequences of any length . By an exhaustive search the conjecture was known to hold up to 25 vertices. We report here that it remains true up to 117 vertices. This is achieved by considering the weaker notion of parity-regular Steinhaus graphs, where all vertex degrees have the same parity. We show that these graphs can be parametrized by an -vector space of dimension approximately and thus constitute an efficiently describable domain where true regular Steinhaus graphs can be searched by computer.

     

Continuous-time Kreiss resolvent condition on infinite-dimensional spaces

Given the infinitesimal generator of a -semigroup on the Banach space which satisfies the Kreiss resolvent condition, i.e., there exists an such that for all complex with positive real part, we show that for general Banach spaces this condition does not give any information on the growth of the associated -semigroup. For Hilbert spaces the situation is less dramatic. In particular, we show that the semigroup can grow at most like . Furthermore, we show that for every there exists an infinitesimal generator satisfying the Kreiss resolvent condition, but whose semigroup grows at least like . As a consequence, we find that for with the standard Euclidian norm the estimate cannot be replaced by a lower power of or .

     

Multigrid methods for the computation of singular solutions and stress intensity factors I: Corner singularities

We consider the Poisson equation with homogeneous Dirichlet boundary condition on a two-dimensional polygonal domain with re-entrant angles. A multigrid method for the computation of singular solutions and stress intensity factors using piecewise linear functions is analyzed. When , the rate of convergence to the singular solution in the energy norm is shown to be , and the rate of convergence to the stress intensity factors is shown to be , where is the largest re-entrant angle of the domain and can be arbitrarily small. The cost of the algorithm is . When , the algorithm can be modified so that the convergence rate to the stress intensity factors is . In this case the maximum error of the multigrid solution over the vertices of the triangulation is shown to be .

     

Analysis of PSLQ, an integer relation finding algorithm

Let be either the real, complex, or quaternion number system and let be the corresponding integers. Let be a vector in . The vector has an integer relation if there exists a vector , , such that . In this paper we define the parameterized integer relation construction algorithm PSLQ, where the parameter can be freely chosen in a certain interval. Beginning with an arbitrary vector , iterations of PSLQ will produce lower bounds on the norm of any possible relation for . Thus PSLQ can be used to prove that there are no relations for of norm less than a given size. Let be the smallest norm of any relation for . For the real and complex case and each fixed parameter in a certain interval, we prove that PSLQ constructs a relation in less than iterations.

     

Double roots of power series and related matters

For a given collection of distinct arguments , multiplicities and a real interval containing zero, we are interested in determining the smallest for which there is a power series with coefficients in , and roots of order respectively. We denote this by . We describe the usual form of the extremal series (we give a sufficient condition which is also necessary when the extremal series possesses at least non-dependent coefficients strictly inside , where is 1 or 2 as is real or complex). We focus particularly on , the size of the smallest double root of a power series lying on a given ray (of interest in connection with the complex analogue of work of Boris Solomyak on the distribution of the random series ). We computed the value of for the rationals in of denominator less than fifty. The smallest value we encountered was . For the one-sided intervals and the corresponding smallest values were and .

     

Comparison theorems of Kolmogorov type and exact values of -widths on Hardy-Sobolev classes

Let be a strip in complex plane. denotes those -periodic, real-valued functions on which are analytic in the strip and satisfy the condition , . Osipenko and Wilderotter obtained the exact values of the Kolmogorov, linear, Gel'fand, and information -widths of in , , and 2-widths of in , , .

In this paper we continue their work. Firstly, we establish a comparison theorem of Kolmogorov type on , from which we get an inequality of Landau-Kolmogorov type. Secondly, we apply these results to determine the exact values of the Gel'fand -width of in , . Finally, we calculate the exact values of Kolmogorov -width, linear -width, and information -width of in , , .