Burgisser, P., University of Zurich, Switzerland

Completeness and Reduction in Algebraic Complexity Theory

2000. Approx. 170 pp., 16 figs.
3-540-66752-0

The theory of NP-completeness is a cornerstone of computational complexity. This monograph
provides a thorough and comprehensive treatment of this concept in the framework of algebraic
complexity theory. Many of the results presented are new and published for the first time. Topics
include: complete treatment of Valiant's algebraic theory of NP-completeness, interrelations with
the classical theory as well as the Blum-Shub-Smale model of computation, questions of structural
complexity, fast evaluation of representations of general linear groups, and complexity of
immanants. The book can be used at the advanced undergraduate or at the beginning graduate
level in either mathematics or computer science.

Keywords: complexity, NP-completeness

Contents: 1. Introduction.- 2. Valiant's Algebraic Model of NP-Completeness.- 3. Some Complete
Families of Polynomials.- 4. Cook's versus Valiant's Hypothesis.- 5. The Structure of Valiant's
Complexity Classes.- 6. Fast Evaluation of Representations of General Linear Groups.- 7. The
Complexity of Immanants.- 8. Separation Results and Future Directions.- References.- List of
Notations.- Index

Series: Algorithms and Computation in Mathematics.VOL. 7

@


Pomerance, C., University of Georgia, Athens, GA, USA
Crandall, R., Portland, OR, USA

Primes

A Computational Perspective

2000. Approx. 350 pp. 13 figs., 10 in color.
0-387-94777-9

Destined to become a definitive textbook conveying the most modern computational ideas about
prime numbers and factoring, this book will stand as an excellent reference for this kind of
computation, and thus be of interest to both educators and researchers. It is also a timely book,
since primes and factoring have reached a certain vogue, partly because of cryptography. The final
chapter focusses on "applications" of prime numbers, incorporating the mathematics of finance, via
quasi-Monte Carlo theory. Historical comments are contained in every chapter.

Contents: The World of Prime Numbers; Fundamental Algorithms; Useful Algorithms; Recognizing
Primes and Composites; Primality Proving; Exponential Factoring Algorithms; Sub-Exponential
Factoring Algorithms; Applications of Prime Numbers; Sparse Bit Matrices; Appendix


Edited by: Werner Haussmann
Kurt Jetter
Manfred Reimer

Advances in Multivariate Approximation

ISBN: 3-527-40236-5
Hardcover
Pages: 334
Published: Oct 1999
Copyright: 1999
Imprint: A Wiley-VCH Publication

This book details the results of the 3rd International Conference on Multivariate Approximation organized by the University of
Dortmund in 1998. Like all volumes in this series, the book provides communications links between the various special groups within
mathematics.

Table of Contents
Problems of Approximation Theory in Discrete Geometry (N. Andreev & V. Yudin).
Dense Vector Spaces of Universal Harmonic Functions (D. Armitage).
Best One-sided L1-Approximation by Harmonic and Subharmonic Functions (D. Armitage & S. Gardiner).
Numerical Stability of Fast Fourier Transforms (G. Baszenski, et al.).
The Saturation Theorem for Box Spline Orthogonal Projection (M. Bes\aaka & K. Dziedziul).
Problem on Monotonic Behaviour of Bernstein Operators (M. Bes\aaka & K. Dziedziul).
Best One-Sided L1-Approximation by Blending Functions (B. Bojanov, et al.).
On the Asymptotics of Points which Maximize Determinants of the Form det [g(\p8vxi - xj\p8v)] (L. Bos & U. Maier).
Besov Regularity for the Stokes Problem (S. Dahlke).
Optimal Periodic Interpolation in Multivariate Periodic Hilbert Spaces (F.-J. Delvos).
Multiscale Modelling from Geopotential Data (W. Freeden, et al.).
On a Family of Orthogonal Wavelets on the Quincunx Grid: Open Regularity Questions (A. Gottscheber & G. Steidl).
Average Case Analysis of Numerical Integration (P. Grabner, et al.).
Cubature Formulas on Spheres (H. König).
Numerical Calculation of Spherical Designs (U. Maier).
On Bivariate Spline Spaces (G. NEnberger & F. Zeilfelder).
Spherical Polynomial Approximations: A Survey (M. Reimer).
Range Restricted Interpolation by Cubic C1-Splines on Clough-Tocher Splits (J. Schmidt).
Some Error Estimates for Periodic Interpolation of Functions from Besov Spaces (W. Sickel & F. Sprengel).
The Uniform Error of Hyperinterpolation on the Sphere (I. Sloan & R. Womersley).
Simultaneous Approximation in the Dirichlet Space (A. Stray).
The Equivalence between Weighted K-Functionals and Moduli of Smoothness on a Simplex and their Applications (P.-C. Xuan).

@


Rio, E., Universite Paris Sud, Orsay, France

Theorie asymptotique des processus aleatoires faiblement dependants

2000. XII, 170 p.
3-540-65979-X

Ces notes sont consacrees aux inegalites et aux theoremeses limites classiques pour les suites de
variables aleatoires absolument regulieres ou fortement melangeantes au sens de Rosenblatt. Le
but poursuivi est de donner des outils techniques pour l'etude des processus faiblement
dependants aux statisticiens ou aux probabilistes travaillant sur ces processus. Nos resultats et
nos preuves sont essentiellement fond sur des inegalites de covariance et des lemmes de
couplage parfois recents, que nous appliquons pour obtenir des theoremes limites classiques tels
que la loi forte des grands nombres avec ou sans vitesses de convergence, le theoreme limite
central et le theoreme limite central fonctionnel pour les sommes partielles normalisees, la loi du
logarithme itere, l'etude des processus empiriques. Enfin nous donnons quelques resultats
theoriques sur les relations entre la vitesse d'ergodicite et la vitesse de melange fort des chaaines
de Markov irreductibles.

Keywords: strongly mixing sequences, absolutely regular sequences, covariance inequalities,
coupling, central limit theorem

Contents: Variance des sommes partielles.- Moments algebriques. Premieres inegalites
exponentielles.- Inegalites maximales et lois fortes.- Le theoreme limite central.- Couplage et
melange.- Inegalites de Fuk-Nagaev, moments d'ordre quelconque.- Fonction de repartition
empirique.- Processus empiriques indexes par des classes de fonctions.- Chaines de Markov
irreductibles.- Annexes.

Series: Mathematiques & Applications.VOL. 31


Chalmond, B., Ecole Normale Superieure de Cachan, France

Elements de modelisation pour l'analyse d'images

2000. XVIII, 331 p.
3-540-66563-3

Cet ouvrage decrit une methodologie et un savoir-faire pour la construction effective de modeles en
analyse d'images. Les taches d'imagerie y sont le plus souvent formalisees comme des problemes
inverses solutionnes dans un cadre Bayesien. Ce livre est organise en 3 parties. Les 2 premieres
decrivent les bases necessaires aux modeles developpes dans la troisieme partie sous la forme d'
d'energie. Ces bases sont les splines et les champs aleatoires. La plupart des modeles sont issus
de projets industriels auxquels l'auteur a participe en radiographie et en controle non desttructif:
suivi de lignes 3D, traitement de degradation en radiographie, reconstruction 3D et tomographie,
mise en correspondance, apprentissage de deformations. De nombreuses illustrations graphiques
accompagnent le texte montrant les performances des modeles proposes.

Keywords: analyse d ' image Problemes inverses approximation spline champs de Markov
radiographie

Contents: Introduction.- A propos de modelisation.- Structure de l'ouvrage.- Modeles splines:
Modeles splines non parametriques.- Modeles splines parametriques.- Modeles auto-associatifs.-
Modeles markoviens: Aspects fondamentaux.- Estimation bayesienne.- Estimation et
optimisation.- Estimation des parametres.- Modelisation en action: Construction de modele.-
Degradation en imagerie.- Detection d'entites filiformes.- Reconstruction et projections.-
Reconstruction regularisee.- Reconstruction avec une seule vue.- Mise en correspondance.

Series: Mathematiques & Applications.VOL. 33