List of Participants

Name: Gerard Awanou
E-mail: gawanou@math.uga.edu
Institute: University of Georgia
Talk Title: Spline approximations of the 3D Navier-Stokes Equations
Talk Abstract: We consider numerical approximations of the 3D Navier-Stokes equations in velocity-pressure formulation. The pressure is eliminated from the equations by using a space of velocity fields which are divergence free. The later is discretized by means of splines of arbitrary degree and arbitrary smoothness. Energy arguments are used to derive the discrete equations satisfied by the equations. The pressure term is computed by solving a Poisson problem with Neumann boundary.
   
Name: Yuliya Babenko
E-mail: yuliya.v.babenko@vanderbilt.edu
Institute: Vanderbilt University
Talk Title: About Kolmogorov problem for (r-1)-monotone functions on the half-line
Talk Abstract: The talk is mainly about Kolmogorov-type inequalities that estimate the $L_q$-norm of the intermediate derivative $x^{(k)}$ of an $(r-1)$-monotone function $x$ defined on a half-line in terms of the $L_p$-norm of $x$ and the $L_\infty$-norm of its derivative $x^{(r)}$. In particular, we will give a solution of the following Kolmogorov problem: find necessary and sufficient conditions such that for any given $M_{0,\infty},M_{k,q},M_{r,\infty}$, there exists an $(r-1)$-monotone function $x$ with $M_{0,\infty}=\left\| x\right\| _{\infty},M_{k,q}=\left\| x^{\left( k\right) }\right\| _{q},M_{r,\infty}=\left\| x^{\left( r\right) }\right\|_{\infty}$.
   
Name: Katherine Balazs
E-mail: kilgokb@auburn.edu
Institute: Department of Mathematics Auburn University
Talk Title: Simultaneous Approximation by Interpolation
Talk Abstract: TBA
   
Name: Victoria Baramidze
E-mail: vbaramid@math.uga.edu
Institute: University of Georgia
Talk Title: Spherical Cubic Interpolating Splines
Talk Abstract: We study several numerical schemes solving a scattered data interpolation problem on a sphere using spherical cubic splines.
   
Name: Sudeshna Basu
E-mail: sbasu@howard.edu
Institute: Howard University
   
Name: Peter Binev
E-mail: binev@math.sc.edu
Institute: University of South Carolina
Talk Title: Local Subdivision on Adaptive Triangulations
Talk Abstract: Representation of surfaces in computer graphics often requires consecutive approximations with different levels of accuracy. A multiresolution representation proves to be very effective, but it usually involves a global (regular) subdivision procedure. On the other hand adaptive local refinement can provide an effective algorithm. Here we propose to use a local refinement of irregular triangulations known as {\it newest vertex bisection}, and an appropriate subdivision scheme based on a quasi-interpolation operator.
   
Name: Kai Bittner
E-mail: bittner@math.umsl.edu
Institute: Department of Mathematics & CS, University of Missouri - St. Louis
Talk Title: Spline Modulation of Sinusoids for Signal Representation
Talk Abstract: We investigate biorthogonal Wilson bases generated by cardinal B-splines. Such bases provide very efficient representations of sinusoids modulated by spline functions. Exact Riesz bound for such bases are given. Furthermore, we derive an explicit formula for the computation of the dual window functions and determine their decay rate.
   
Name: Derek Bruff
E-mail: bruff@math.vanderbilt.edu
Institute: Vanderbilt University
   
Name: Carlos Cabrelli
E-mail: cabrelli@math.gatech.edu
Institute: University of Buenos Aires and Georgia Institute of Technology
   
Name: Steven Damelin
E-mail: damelin@gasou.edu
Institute: Georgia Southern University
Talk Title: Distribution of general interpolation arrays for exponential weights
Talk Abstract: We discuss the distribution of general interpolation arrays for a large class of exponentially decaying weights on $(-1,1)$ and the line.
   
Name: Arthur Danielyan
E-mail: adaniely@pegasus.cc.ucf.edu
Institute: University of Central Florida
Talk Title: On a Localization Theorem for Approximation in Complex Domain
Talk Abstract: We consider the problem of representation of Baire first class functions on the compact sets of the complex plane by convergent sequences of rational functions and prove a corresponding localization theorem. We also present several open problems.
   
Name: Frank Deutsch
E-mail: deutsch@math.psu.edu
Institute: Penn State University
   
Name: Narendra Kumar Govil
E-mail: govilnk@auburn.edu
Institute: Auburn University, Auburn, AL 36849
Talk Title: On Growth of Polynomials
Talk Abstract: Let $p(z) = \sum_{\nu=0}^n a_\nu z^\nu$ be a polynomial of degree $n$, $M(p,R) = \max\limits_{|z| = R \ge 1} |p(z)|$ and let $\|p\| = \max\limits_{|z| = 1} |p(z)|$. If $p(z) \neq 0$ in $|z| < 1$, then according to a well known result of Ankeny and Rivlin, $M(p,R) \le \left( \frac{R^n + 1}{2} \right) \|p\| $ for $R \ge 1$. In this talk, we would be presenting some results that generalize and sharpen this and some other related inequalities.
   
Name: Doug Hardin
E-mail: hardin@math.vanderbilt.edu
Institute: Vanderbilt University
Talk Title: The matrix Fej\'er-Riesz Theorem and the construction of local orthonormal shift-invariant bases.
Talk Abstract: We present a constructive proof of the matrix-valued Fej\'er-Riesz Theorem with real coefficients and characterize when a univariate shift-invariant space has a local orthonormal shift-invariant basis.
   
Name: Wenjie He
E-mail: he@neptune.cs.umsl.edu
Institute: Dept. of Math. and Computer Science, Univ. of Missouri, St. Louis
Talk Title: Compactly supported symmetric tight wavelet frames
Talk Abstract: We consider compactly supported tight frames $\Psi=\{ \psi^1, \ldots, \psi^N \}$ for $L^2(\RR^2)$ that associate with some compactly supported refinable functions. We introduce an algorithmic approach for constructing tight frame generators. This approach features a recipe for selecting the columns of the Kronecker product of the orthonormal matrix extensions of the univariate two-scale symbols in order to formulate the multivariate orthonormal matrix extension of the multivariate two-scale symbol. We pay particular attention to the case that each tight frame generator is either symmetric or antisymmetric.
   
Name: Christopher Heil
E-mail: heil@math.gatech.edu
Institute: Georgia Tech
   
Name: Don Hong
E-mail: hong@math.vanderbilt.edu, hong@etsu.edu
Institute: E T S U / Vanderbilt University
   
Name: Bruce Kessler
E-mail: bruce.kessler@wku.edu
Institute: Western Kentucky University
Talk Title: Optimal Prefilters for a DGH 2-D Orthogonal Scaling Vector
Talk Abstract: In the case where a single scaling func tion is used to represent a signal, the discrete data can be used as scaling fun ction coefficients (the identity prefilter) and polynomial degree of data (up to the approximation order of the $V_0$ space) is preserved. This is not generall y true, however, with scaling vectors of multiplicity $r>1$. This talk will pre sent a generalization of work by Hardin and Roach (1998) to find prefilters for a dilation-3 orthogonal scaling vector (created by Donovan, Geronimo, and Hardin (1995) and defined over a uniform triangular lattice of ${\bf R}^2$) that will preserve up to linea r approximation order and also preserve the norm of the data. Examples of the effectiveness of the prefilters will be provided.
   
Name: Theodore Kilgore
E-mail: kilgota@banach.math.auburn.edu
Institute: Auburn University
Talk Title: Simultaneous approximation on the line and on the half-line for certain rational functions
Talk Abstract: Some approximation properties on $(-\infty,\infty)$ by rational functions with denominator $(1+x^2)^{-n}$ will be addressed. Analogous results will be presented, too, for approximation on $[0,\infty)$ by rational functions with denominator $(1+t)^{-n}$.
   
Name: Tom Kunkle
E-mail: kunklet@cofc.edu
Institute: College of Charleston
Talk Title: Favard's Interpolation Problem in Several Variables
Talk Abstract: \def\RR{{{\rm I}\kern-.2em {\rm R}}} Let $n$ and $d$ be natural numbers and consider the following problem. Construct a smooth interpolant $F$ to function values $f$ given at points $m$ in $\RR^d$, where $F$ depends locally and linearly on $f$, and $F$'s derivatives of total degree $n$ are bounded by a constant $C$ times the corresponding divided differences of $f$. Here $C$ may depend on $d$ and $n$, but must be independent of $f$ and $m$. \par Favard gave an optimal solution to the problem in case $d=1$ by bounding each of two consecutive derivatives by the corresponding divided differences and allowing $m$ to be any discrete set in $\RR$. Such freedom would not be possible in the multivariate setting. In fact, if $m$ is a tensor product grid in $\RR^d$, i.e., the Cartesian product of $d$ increasing sequences of real numbers, then such an interpolation scheme exists if and only if $m$ has uniform spacing in each of the coordinate directions. When $m$ is a subset of a tensor product grid with uniform spacing, the interpolation problem may or may not have a solution, depending on the geometry of the subset.
   
Name: John Lavery
E-mail: lavery@aro.arl.army.mil
Institute: Army Research Office
Talk Title: Shape-preserving approximation of multiscale univariate data by cubic $L_1$ splines
Talk Abstract: Spline fits are calculated by minimizing a data fitting functional over a manifold of splines. Smoothing splines are calculated by minimizing a linear combination of the data fitting functional and the interpolating spline functional. For multiscale data, that is, data with abrupt changes in magnitude and/or spacing, currently available spline fits and smoothing splines and spline fits typically have extraneous oscillation. In this paper, a class of cubic $L_1$ spline fits that do not have extraneous oscillation is investigated. Cubic $L_1$ spline fits are compared with the conventional $L_2$ cubic spline fits and with cubic $L_1$ and $L_2$ smoothing splines. Cubic $L_1$ spline fits are calculated by a Lagrange-multiplier-based primal affine (interior point) algorithm. Computational results for three data sets are given. $L_1$ spline fits preserve shape well. $L_1$ spline fits are computationally more expensive than $L_1$ smoothing splines but require only half the storage and do not depend on a parameter that the user must choose.
   
Name: Jian-Guo Liu
E-mail: jliu@math.umd.edu
Institute: University of Maryland
Talk Title: Efficient Numerical Methods for Incompressible Flow
Talk Abstract: In this talk, I will review a class of numerical methods for 2D and 3D unsteady incompressible Navier-Stokes equations for large Reynolds number based on local vorticity boundary conditions or local pressure boundary conditions. In each case a convectionally stable explicit time stepping is used for the momentum equation and the incompressibility constraint is realized by a derived Poisson equation. The local vorticity boundary conditions and local pressure boundary conditions are used so that the dynamic and kinematic equations are decoupled. The result in a class of very efficient Naver-Stokes solver. At each time step, only one Poisson equation needs to be solved to achieve 2nd order accuracy. With a modest increase in the computational cost, higher order approximations are possible using compact finite difference approximations, finite element approximations, or spectral approximations. A patch mesh refinement can be easily incorporated into the schemes to resolve highly unstable boundary layers. Various numerical experiments will be presented to demonstrate the accuracy and efficiency of this class of methods, including the flow past cylinder at the Reynolds numeber up to 200,000 and an air bubble rising water.
   
Name: Doron Lubinsky
E-mail: lubinsky@math.gatech.edu
Institute: Georgia Institute of Technology
Talk Title: Rogers-Ramanujan and the Baker-Gammel-Wills (Pade) Conjecture
Talk Abstract: In 1961, Baker, Gammel and Wills conjectured that a subsequence of the diagonal Pade sequence to a function f meromorphic in the unit ball, converges away from the poles. We showed recently that the Rogers-Ramanujan continued fraction provides a counterexample. We discuss this and subsequent developments. No Pade background is assumed.
   
Name: Hrushikesh Mhaskar
E-mail: hmhaska@calstatela.edu
Institute: California State University, Los Angeles
Talk Title: Bit representation of band-dominant functions on the sphere
Talk Abstract: A band-dominant function on the Euclidean sphere embedded in ${\bf R}^{q+1}$ is the restriction to this sphere of an entire function of $q+1$ complex variables having a finite exponential type in each of its variables. We develop a method to represent such a function using finitely many bits, using the values of the function at scattered sites on the sphere. The number of bits required in our representation is asymptotically the same as the metric entropy of the class of such functions with respect to any of the $L^p$ norms on the sphere.
   
Name: Erwin Mina
E-mail: emina@math.vanderbilt.edu
Institute: Vanderbilt University
   
Name: Ram Mohapatra
E-mail: ramm@pegasus.cc.ucf.edu
Institute: University of Central Florida
Talk Title: On Fractional Order Derivatives of Trigonometric Polynomials
Talk Abstract: In this paper we consider Weyl derivatives of fractional order and obtain necessary and sufficvient conditions for a Szego type inequality to hold. We also extend to fractional order an estimate for the derivative of a self inversive polynomial.
   
Name: Ursula Molter
E-mail: umolter@math.gatech.edu
Institute: University of Buenos Aires and Georgia Institute of Technology
   
Name: Mike Neamtu
E-mail: neamtu@math.vanderbilt.edu
Institute: Vanderbilt University
Talk Title: Multivariate Splines - Recent Progress
Talk Abstract: In this survey talk I will discuss recent developments in the theory of multivariate splines constructed via simplex splines.
   
Name: David Roach
E-mail: david.roach@murraystate.edu
Institute: Murray State University
Talk Title: Explicit Parameterizations of Wavelets
Talk Abstract: In this talk, we give a complete and simple parameterization for the length six one-dimensional filters which satisfy the necessary conditions for orthogonality. This parameterization includes all compactly supported univariate scaling functions contained within the interval $[0,5]$ using dilation factor 2. These formulae are a convenient way to generate a continuum of wavelets with varying approximation properties ranging from scaling functions which only reproduce constants to ones which reproduce linears and quadratics. In addition, solutions for the length eight and ten filters are given where the parameters have transcendental constraints. We conclude the talk with some interesting numerical experiments presenting a parameterized wavelet of length six which performs better than the standard wavelets of Haar, D4, and D6 in an image compression scheme on a variety of images.
   
Name: Larry Schumaker
E-mail: s@mars.cas.vanderbilt.edu
Institute: Vanderbilt university
Talk Title: Surface Compression Using $C^1$ Cubic Splines on Triangulated Quadrangulations
Talk Abstract: The space of $C^1$ cubic splines defined on triangulated quadrangulations is used to construct a Faber-type multiresolution interpolation method for approximating and compressing surfaces defined by bivariate functions.
   
Name Boris Shekhtman
E-mail boris@math.usf.edu
Institution University of South Florida
Talk Title Interpolation by Polynomials in Several Variables.
Abstract We discuss the dimension of uniformly interpolating subspaces in several variables.
   
Name: Tatyana Sorokina
E-mail: sorokina@math.vanderbilt.edu
Institute: Vanderbilt University
Talk Title: Quintic Spline Interpolation on Type-4 Tetrahedral Partitions
Talk Abstract: Given a rectangular box $B$, we define the associated type-4 tetrahedral partition $\tri$ by first splitting $B$ into suboxes, then splitting each subbox into 5 tetrahedra. The partition can be regarded as a cross-cut partition which is created by cutting $B$ using planes with four different orientations. We construct a method for interpolating Hermite data at the vertices of $\tri$ using a space of $C^1$ quintic supersplines. We prove that the method has full approximation power, and also present several numerical results.
   
Name: Joachim Stoeckler
E-mail: joachim.stoeckler@math.uni-dortmund.de
Institute: University of Dortmund
Talk Title: Spline frames on the real line and bounded intervals
Talk Abstract: We present new concepts for the construction of tight frames on the real line with maximum vanishing moments. Similar concepts can be introduced for the construction of tight frames on bounded intervals. As an example we study frames that are linear combinations of of cubic B-splines with arbitrary knots.
   
Name: Paul Wenston
E-mail: Paul@math.uga.edu
Institute: University of Georgia
Talk Title: Bivarate C^1 Cubic Splines for Exterior Biharmonic Problems
Talk Abstract: C^1 cubic splines and the infinite element method are used to solve the both the homogeneous and nonhomogenous biharmonic equation exterior to a bounded domain.
   
Name: Frank Zeilfelder
E-mail: zeilfeld@euklid.math.uni-mannheim.de
Institute: University of Mannheim, Germany
Talk Title: Local Lagrange Interpolation by Bivariate Splines
Talk Abstract: There exists a vast literature on local Hermite interpolation by smooth splines on triangulations. On the other hand, since recently no results of this type were known for Lagrange interpolation. We develop the first local Lagrange interpolation methods for $C^1$ splines on triangulations and triangulated quadrangulations. The methods are based on a suitable coloring of the triangles and quadrilaterals, respectively. The Lagrange interpolating splines yield optimal approximation order and can be computed with linear complexity. We give numerical tests which demonstrate the efficiency of the methods. The results were obtained by joint work with G’nther N’rnberger and Larry L. Schumaker.