Mellin Transforms of Multivariate Rational Functions

This paper deals with Mellin transforms of rational functions g/f in several variables. We prove that the polar set of such a Mellin transform consists of finitely many families of parallel hyperplanes, with all planes in each such family being integral translates of a specific facial hyperplane of the Newton polytope of the denominator f. The Mellin transform is naturally related to the so-called coamoeba $\mathcal{A}'_{f}:=\mathrm{Arg}(Z_{f})$ , where Z _f is the zero locus of f and Arg denotes the mapping that takes each coordinate to its argument. In fact, each connected component of the complement of the coamoeba $\mathcal{A}'_{f}$ gives rise to a different Mellin transform. The dependence of the Mellin transform on the coefficients of f, and the relation to the theory of A-hypergeometric functions is also discussed in the paper.     

Harmonic Analysis of Neural Networks

It is known that superpositions of ridge functions (single hidden-layer feedforward neural networks) may give good approximations to certain kinds of multivariate functions. It remains unclear, however, how to effectively obtain such approximations. In this paper, we use ideas from harmonic analysis to attack this question. We introduce a special admissibility condition for neural activation functions. The new condition is not satisfied by the sigmoid activation in current use by the neural networks community; instead, our condition requires that the neural activation function be oscillatory. Using an admissible neuron we construct linear transforms which represent quite general functionsfas a superposition of ridge functions. We develop

• • • a continuous transform which satisfies a Parseval-like relation;
• • • a discrete transform which satisfies frame bounds.

Both transforms representfin a stable and effective way. The discrete transform is more challenging to construct and involves an interesting new discretization of time–frequency–direction space in order to obtain frame bounds for functions inL²(A) whereAis a compact set of Rⁿ. Ideas underlying these representations are related to Littlewood–Paley theory, wavelet analysis, and group representation theory.     

Injectivity of rotation invariant windowed Radon transforms

We consider rotation invariant windowed Radon transforms that integrate a function over hyperplanes by using a radial weight (called window). T. Quinto proved their injectivity for square integrable functions of compact support. This cannot be extended in general. Actually, when the Laplace transform of the window has a zero with positive real part δ, the windowed Radon transform is not injective on functions with a Gaussian decay at infinity, depending on δ. Nevertheless, we give conditions on the window that imply injectivity of the windowed Radon transform on functions with a more rapid decay than any Gaussian function.     

The complexity probabilistic quasi-metric space

We introduce and study a probabilistic quasi-metric on the set of complexity functions, which provides an efficient framework to measure the distance from a complexity function f to another one g in the case that f is asymptotically more efficient than g. In this context we also obtain a version of the Banach fixed point theorem which allows us to show that the functionals associated both to Divide and Conquer algorithms and Quicksort algorithms have a unique fixed point.     

Algorithms for projecting a point onto a level surface of a continuous function on a compact set

Given an equation f(x) = 0, the problem of finding its solution nearest to a given point is considered. In contrast to the authors’ previous works dealing with this problem, exact algorithms are proposed assuming that the function f is continuous on a compact set. The convergence of the algorithms is proved, and their performance is illustrated with test examples.     

Some Results on Condition Numbers in Convex Multiobjective Optimization

Various notions of condition numbers are used to study some sensitivity aspects of scalar optimization problems. The aim of this paper is to introduce a notion of condition number to study the case of a multiobjective optimization problem defined via m convex C ^1,1 objective functions on a given closed ball in ? ⁿ . Two approaches are proposed: the first one adopts a local point of view around a given solution point, whereas the second one considers the solution set as a whole. A comparison between the two notions of well-conditioned problem is developed. We underline that both the condition numbers introduced in the present work reduce to the same condition number proposed by Zolezzi in 2003, in the special case of the scalar optimization problem considered there. A pseudodistance between functions is defined such that the condition number provides an upper bound on how far from a well–conditioned function f a perturbed function g can be chosen in order that g is well–conditioned too. For both the local and the global approach an extension of classical Eckart–Young distance theorem is proved, even if only a special class of perturbations is considered.     

Continuous shearlet frames and resolution of the wavefront set

In recent years directional multiscale transformations like the curvelet- or shearlet transformation have gained considerable attention. The reason for this is that these transforms are??unlike more traditional transforms like wavelets??able to efficiently handle data with features along edges. The main result in Kutyniok and Labate (Trans. Am. Math. Soc. 361:2719?C2754, 2009) confirming this property for shearlets is due to Kutyniok and Labate where it is shown that for very special functions ?? with frequency support in a compact conical wegde the decay rate of the shearlet coefficients of a tempered distribution f with respect to the shearlet ?? can resolve the wavefront set of f. We demonstrate that the same result can be verified under much weaker assumptions on ??, namely to possess sufficiently many anisotropic vanishing moments. We also show how to build frames for ${L^2(\mathbb{R}^2)}$ from any such function. To prove our statements we develop a new approach based on an adaption of the Radon transform to the shearlet structure.     

On Chebyshev functions and Klee functions

Associated to a lower semicontinuous function, one can define its proximal mapping and farthest mapping. The function is called Chebyshev (Klee) if its proximal mapping (farthest mapping) is single-valued everywhere. We show that the function f is 1/λ-hypoconvex if its proximal mapping Pλf is single-valued. When the function f is bounded below, and Pλf is single-valued for every λ>0, the function must be convex. Similarly, we show that the function f is 1/μ-strongly convex if the farthest mapping Qμf is single-valued. When the function is the indicator function of a set, this recovers the well-known Chebyshev problem and Klee problem in Rn. We also give an example illustrating that a continuous proximal mapping (farthest mapping) needs not be locally Lipschitz, which answers one open question by Hare and Poliquin.     

Variations on a theme of Boole and Stein-Weiss

We give an alternative proof of a theorem of Stein and Weiss: The distribution function of the Hilbert transform of a characteristic function of a set E only depends on the Lebesgue measure |E| of such a set. We exploit a rational change of variable of the type used by George Boole in his paper “On the comparison of transcendents, with certain applications to the theory of definite integrals” together with the observation that if two functions f and g have the same Lp norm in a range of exponents p₁<p<p₂ then their distribution functions coincide.     

Fourier transforms and bent functions on finite groups

Let G be a finite nonabelian group. Bent functions on G are defined by the Fourier transforms at irreducible representations of G. We introduce a dual basis \({\widehat{G}}\), consisting of functions on G determined by its unitary irreducible representations, that will play a role similar to the dual group of a finite abelian group. Then we define the Fourier transforms as functions on \({\widehat{G}}\), and obtain characterizations of a bent function by its Fourier transforms (as functions on \({\widehat{G}}\)). For a function f from G to another finite group, we define a dual function \({\widetilde{f}}\) on \({\widehat{G}}\), and characterize the nonlinearity of f by its dual function \({\widetilde{f}}\). Some known results are direct consequences. Constructions of bent functions and perfect nonlinear functions are also presented.     

Analytic solutions of a nonlinear iterative equation near neutral fixed points and poles

In this paper existence of analytic solutions of a nonlinear iterative equations is studied when given functions are all analytic and when given functions have poles. As well as in many previous works, we reduce this problem to finding analytic solutions of a functional equation without iteration of the unknown function f. For technical reasons, in previous works an indeterminate constant related to the eigenvalue of the linearized f at its fixed point O is required to fulfill the Diophantine condition that O is an irrationally neutral fixed point of f. In this paper the case of rationally neutral fixed points is also discussed, where the Diophantine condition is not required.     

Best approximation by linear combinations of characteristic functions of half-spaces

It is shown that for any positive integer n and any function f in with p∈[1,∞) there exist n half-spaces such that f has a best approximation by a linear combination of their characteristic functions. Further, any sequence of linear combinations of n half-space characteristic functions converging in distance to the best approximation distance has a subsequence converging to a best approximation, i.e., the set of such n-fold linear combinations is an approximatively compact set.     

Limits of inverse limits

We study the following problem: if a sequence of graphs of upper semi-continuous set valued functions fn converges to the graph of a function f, is it true that the sequence of corresponding inverse limits obtained from fn converges to the inverse limit obtained from f?     

Separation and probability of correct classification among two or more distributions

A function on theK-fold product of a set in normed vector space will be called a separation measurement if, for any collection ofK points, the function is bounded below and above, respectively, by maximum and total distance between pairs of points in the collection. Separation measurements are relavent toK-sample hypothesis testing and also to discrimination amongK classes, and several examples are given. In particular, ordinaryL ₁ distance between integrable functions can be generalized to a non-pairwise separation measurement for densitiesf ₁,f ₂,...,f _K inL ₁[μ]; and this separation is a linear transform of the optimal discriminant's probability of correct classification. This research was supported by grant A8044 from the National Research Council of Canada.     

Generalized Splines for Radon Transform on Compact Lie Groups with Applications to Crystallography

The Radon transform $\mathcal{R}f$ of functions f on SO(3) has recently been applied extensively in texture analysis, i.e. the analysis of preferred crystallographic orientation. In practice one has to determine the orientation probability density function f∈L ₂(SO(3)) from $\mathcal{R}f\in L_{2}(S^{2}\times S^{2})$ which is known only on a discrete set of points. Since one has only partial information about $\mathcal{R}f$ the inversion of the Radon transform becomes an ill-posed inverse problem. Motivated by this problem we define a new notion of the Radon transform $\mathcal{R}f$ of functions f on general compact Lie groups and introduce two approximate inversion algorithms which utilize our previously developed generalized variational splines on manifolds. Our new algorithms fit very well to the application of Radon transform on SO(3) to texture analysis.     

The realizability problem for Golovach-type functions

The problem mentioned in the title is stated as follows. Consider a function f with some necessary properties of the Golovach function, namely, a piecewise constant nonincreasing right continuous function defined on the set of nonnegative real numbers and taking integer values such that this function is identically equal to 1 at sufficiently large argument values. The problem of realizing the function f in a class $\mathbb{G}$ of topological graphs is to find a graph $G \in \mathbb{G}$ such that its Golovach functions coincides with f. Examples of realization of some functions possessing the properties mentioned above are considered. In the simplest case, all graphs for which the function can be realized are described. For less trivial examples, realizability criteria for functions with the properties of the Golovach function in the class of trees and in the class of trees with given edge search number which have the least number of edges are presented.     

Convergence of interpolation in polynomial Chebyshev approximation

The quality of a polynomial approximation on an interval to a functionf is considered as a function of its points of interpolation. Iff satisfies a Lipschitz condition of order 1, the quality depends linearly on the distance of the points of interpolation from an optimal interpolating point set: further restrictions onf still give only linear dependence. This suggests that algorithms based on interpolation are inferior to algorithms based on error extrema (such as the Remes algorithm).     

Basic properties of wavelets

A wavelet multiplier is a function whose product with the Fourier transform of a wavelet is the Fourier transform of a wavelet. We characterize the wavelet multipliers, as well as the scaling function multipliers and low pass filter multipliers. We then prove that if the set of all wavelet multipliers acts on the set of all MRA wavelets, the orbits are the sets of all MRA wavelets whose Fourier transforms have equal absolute values, and these are also equal to the sets, of all MRA wavelets with the corresponding scaling functions having the same absolute values of their Fourier transforms. As an application of these techniques, we prove that the set of MRA wavelets is arcwise connected in L²(R). Dedicated to Eugene Fabes The Wutam Consortium     

Row-Strict Quasisymmetric Schur Functions

Haglund, Luoto, Mason, and van Willigenburg introduced a basis for quasisymmetric functions, called the quasisymmetric Schur function basis, generated combinatorially through fillings of composition diagrams in much the same way as Schur functions are generated through reverse column-strict tableaux. We introduce a new basis for quasisymmetric functions, called the row-strict quasisymmetric Schur function basis, generated combinatorially through fillings of composition diagrams in much the same way as quasisymmetic Schur functions are generated through fillings of composition diagrams. We describe the relationship between this new basis and other known bases for quasisymmetric functions, as well as its relationship to Schur polynomials. We obtain a refinement of the omega transform operator as a result of these relationships.     

On the Laplace transforms of retarded,Lorentz-invariant functions

Let ø(t) (t ∈ $\text{R}$ⁿ) be a retarded, Lorentz-invariant function which satisfies, in addition, condition (c). We call “R” the family of such functions. Let f(z) be the Laplace transform of ø(t) ∈ $\text{R}$. We prove (Theorem 1) that f(z) can be expressed as a K-transform (formula (I, 2; 1)). We apply this formula to evaluate several Laplace transforms. We show that it affords simple proofs of important known results. Formula (I, 2; 1) is an effective complement to L. Schwartz' method of evaluating Fourier transforms via Laplace transforms (“Théorie des distributions,” p. 264, Hermann, Paris, 1966). We think this is the most useful application of our formula.