Converse and Smoothness Theorems for Erdo&#x030B;s Weights inL_(0<p?∞)

We prove converse and smoothness theorems of polynomial approximation in weightedL_pspaces with norm ‖fW‖_Lp()(0<p?∞) for Erdo&#x030B;s weights on the real line. In particular we prove characterization theorems involving realization functionals and thereby establish some interesting properties of our weighted modulus of continuity.     

Representation and approximation of multivariate functions with mixed smoothness by hyperbolic wavelets

In this paper, we study the representation theorems of multivariate functions with mixed smoothness by wavelet basis formed by tensor products of univariate wavelets, we also study the best approximation in the metric for some function classes with mixed smoothness by hyperbolic wavelets and obtain some asymptotic estimates of approximating order.     

A new characterization of projections of quadrics in finite projective spaces of even characteristic

We will classify, up to linear representations, all geometries fully embedded in an affine space with the property that for every antiflag {p,L} of the geometry there are either 0, α, or q lines through p intersecting L. An example of such a geometry with α=2 is the following well known geometry . Let Q_n+1 be a nonsingular quadric in a finite projective space , n≥3, q even. We project Q_n+1 from a point r∉Q_n+1, distinct from its nucleus if n+1 is even, on a hyperplane not through r. This yields a partial linear space whose points are the points p of , such that the line 〈p,r〉 is a secant to Q_n+1, and whose lines are the lines of which contain q such points. This geometry is fully embedded in an affine subspace of and satisfies the antiflag property mentioned. As a result of our classification theorem we will give a new characterization theorem of this geometry.     

Approximation and entropy numbers in Besov spaces of generalized smoothness

We determine the exact asymptotic behaviour of entropy and approximation numbers of the limiting restriction operator , defined by J(f)=f|_Ω. Here Ω is a non-empty bounded domain in , ψ is an increasing slowly varying function, , and is the Besov space of generalized smoothness given by the function t^sψ(t). Our results improve and extend those established by Leopold [Embeddings and entropy numbers in Besov spaces of generalized smoothness, in: Function Spaces, Lecture Notes in Pure and Applied Mathematics, vol. 213, Marcel Dekker, New York, 2000, pp. 323–336].     

Bounded Functions in Möbius Invariant Dirichlet Spaces

Forp∈(0, 1), letQ_p(Q_{p, 0}) be the space of analytic functionsfon the unit diskΔwith sup_w∈Δ ‖f°?_w‖_p<∞ (lim_|w|→1 ‖f°?_w‖_p=0), where ‖·‖_pmeans the weighted Dirichlet norm and?_wis the Möbius map ofΔonto itself with?_w(0)=w. In this paper, we prove the Corona theorem for the algebraQ_p∩H^∞(Q_{p, 0}∩H^∞); then we provide a Fefferman–Stein type decomposition forQ_p(Q_{p, 0}), and finally we describe the interpolating sequences forQ_p∩H^∞(Q_{p, 0}∩H^∞)).     

Pósa-condition and nowhere-zero 3-flows

Let G be a simple graph on n vertices and π(G)=(d₁,d₂,…,d_n) be the degree sequence of G, where n≥3 and d₁≤d₂≤?≤d_n. The classical Pósa’s theorem states that if d_m≥m+1 for and d_m+1≥m+1 for n being odd and , then G is Hamiltonian, which implies that G admits a nowhere-zero 4-flow. In this paper, we further show that if G satisfies the Pósa-condition that d_m≥m+1 for and d_m+1≥m+1 for n being odd and , then G has no nowhere-zero 3-flow if and only if G is one of seven completely described graphs.     

Gradient estimates and Jackson’s theorem in spaces related to measures

Jackson’s theorem is established in a new kind of holomorphic function space Q_μ related to measures in any starlike circular domain in . Particularly, the result covers many spaces including BMOA, Q_p, Q_K, and F(p,q,s) spaces in the unit ball of . Moreover, we construct integral operators which give pointwise estimates for the gradient of the difference in terms of the gradient on the boundary. The gradient estimates are independent of the measures in question and give rise to Jackson’s theorem.     

The Diametric Theorem in Hamming Spaces—Optimal Anticodes

For a Hamming space (ⁿ_α, d_H), the set ofn-length words over the alphabet _α = {0, 1,…,α − 1} endowed with the distanced_H, which for two wordsxⁿ = (x₁,…,x_n),yⁿ = (y₁,…,y_n) ∈ ⁿ_αcounts the number of different components, we determine the maximal cardinality of subsets with a prescribed diameterdor, in another language, anticodes with distanced. We refer to the result as the diametric theorem.In a sense anticodes are dual to codes, which have a prescribedlowerbound on the pairwise distance. It is a hopeless task to determine their maximal sizes exactly.We find it remarkable that the diametric theorem (for arbitrary α) can be derived from our recent complete intersection theorem, which can be viewed as a diametric theorem (for α = 2) in the restricted case, where alln-length words considered have exactlykones.     

Counting lattice chains and Delannoy paths in higher dimensions

Lattice chains and Delannoy paths represent two different ways to progress through a lattice. We use elementary combinatorial arguments to derive new expressions for the number of chains and the number of Delannoy paths in a lattice of arbitrary finite dimension. Specifically, fix nonnegative integers n₁,…,n_d, and let L denote the lattice of points (a₁,…,a_d)∈Z^d that satisfy 0≤a_i≤n_i for 1≤i≤d. We prove that the number of chains in L is given by where . We also show that the number of Delannoy paths in L equals Setting n_i=n (for all i) in these expressions yields a new proof of a recent result of Duchi and Sulanke [9] relating the total number of chains to the central Delannoy numbers. We also give a novel derivation of the generating functions for these numbers in arbitrary dimension.     

Increasing the polynomial reproduction of a quasi-interpolation operator

Quasi-interpolation is an important tool, used both in theory and in practice, for the approximation of smooth functions from univariate or multivariate spaces which contain , the d-variate polynomials of degree ≤m. In particular, the reproduction of Π_m leads to an approximation order of m+1. Prominent examples include Lagrange and Bernstein type approximations by polynomials, the orthogonal projection onto Π_m for some inner product, finite element methods of precision m, and multivariate spline approximations based on macroelements or the translates of a single spline.For such a quasi-interpolation operator L which reproduces and any r≥0, we give an explicit construction of a quasi-interpolant which reproduces Π_m+r, together with an integral error formula which involves only the (m+r+1)th derivative of the function approximated. The operator is defined on functions with r additional orders of smoothness than those on which L is defined. This very general construction holds in all dimensions d. A number of representative examples are considered.     

Estimates of the coefficients of the Jacobi expansion by measures of smoothness

For expansion by Jacobi polynomials we relate smoothness given by appropriate K-functionals in Lp, 1?p?2, to estimates on the coefficients in the ?q form. As a corollary for 1<p?2, and an the coefficients of the Legendre expansion of f∈Lp[−1,1], we obtain the estimate

     

On the Value of Partial Information for Learning from Examples

The PAC model of learning and its extension to real valued function classes provides a well-accepted theoretical framework for representing the problem of learning a target functiong(x) using a random sample {(x_i,g(x_i))}. Based on the uniform strong law of large numbers the PAC model establishes the sample complexity, i.e., the sample sizemwhich is sufficient for accurately estimating the target function to within high confidence. Often, in addition to a random sample, some form of prior knowledge is available about the target. It is intuitive that increasing the amount of information should have the same effect on the error as increasing the sample size. But quantitatively how does the rate of error with respect to increasing information compare to the rate of error with increasing sample size? To answer this we consider a new approach based on a combination of information-based complexity of Traubet al.and Vapnik–Chervonenkis (VC) theory. In contrast to VC-theory where function classes of finite pseudo-dimension are used only for statistical-based estimation, we let such classes play a dual role of functional estimation as well as approximation. This is captured in a newly introduced quantity, ρ_d(), which represents a nonlinear width of a function class . We then extend the notion of thenth minimal radius of information and define a quantityI_{n, d}() which measures the minimal approximation error of the worst-case targetg∈ by the family of function classes having pseudo-dimensiondgiven partial information ongconsisting of values taken bynlinear operators. The error rates are calculated which leads to a quantitative notion of the value of partial information for the paradigm of learning from examples.     

Uniformly nonsquare Banach spaces have the fixed point property for nonexpansive mappings

It is shown that if the modulus Γ_X of nearly uniform smoothness of a reflexive Banach space satisfies , then every bounded closed convex subset of X has the fixed point property for nonexpansive mappings. In particular, uniformly nonsquare Banach spaces have this property since they are properly included in this class of spaces. This answers a long-standing question in the theory.     

Estimates of positive linear operators in terms of second-order moduli

We estimate the constants related with the direct result for positive linear operators which preserves linear functions. The estimates are presented for the modulus of smoothness , where the weight φ is a function such that φ² is concave.     

Rotations of Periodic Strings and Short Superstrings

This paper presents two simple approximation algorithms for the shortest superstring problem with approximation ratios 2 ( ≈ 2.67) and( ≈ 2.596). The framework of our improved algorithms is similar to that of previous algorithms in the sense that they construct a superstring by computing some optimal cycle covers on the distance graph of the given strings and then break and merge the cycles to finally obtain a Hamiltonian path, but we make use of new bounds on the overlap between two strings. We prove that for each periodic semiinfinite string α = a₁a₂··· of periodq, there exists an integerk, such that forany(finite) stringsof periodpwhich isinequivalentto α, the overlap betweensand therotationα[k] = a_ka_{k + 1}··· is at mostp + q. Moreover, ifp ≤ q, then the overlap betweensand α[k] is not larger than (p + q). The bounds are tight. In the previous shortest superstring algorithmsp + qwas used as the standard (tight) bound on overlap between two strings with periodspandq.     

On some geometric parameters in Banach spaces

Let X be a normed linear space and be the unit sphere of X. Let , , and J(X)=sup{‖x+y‖∧‖x−y‖}, x and y∈S(X) be the modulus of convexity, the modulus of smoothness, and the modulus of squareness of X, respectively. Let . In this paper we proved some sufficient conditions on δ(?), ρX(?), J(X), E(X), and , where the supremum is taken over all the weakly null sequence xn in X and all the elements x of X for the uniform normal structure.