NA Digest Sunday, May 18, 2003 Volume 03 : Issue 20

Today's Editor:
Cleve Moler
The MathWorks, Inc.

Submissions for NA Digest:

Mail to

Information via e-mail about NA-NET: Mail to


From: Walter Gander <>
Date: Thu, 15 May 2003 12:09:19 +0200 (MEST)
Subject: Ambros Speiser

Prof. Ambros Speiser, Swiss computer pioneer, passed away on Sat May
10, 2003 at the age of 81, following a stroke the day before. Ambros
Speiser was collaborator together with Heinz Rutishauser in the group
of Eduard Stiefel. Stiefel founded the Institute of Applied
Mathematics at ETH in 1948. Speiser and Rutishauser studied the
development of computers in USA from October 1948 till the end of 1949
at Harvard University and at the Institute for Advanced Studies in
Princeton. In that time Stiefel rented the Z4 from Konrad Zuse for
ETH. The Z4, the first computer in continental Europe after the war,
was operational from 1950 to 1954 at ETH. The Z4 was improved in many
aspects by Speiser. Speiser was also the chief engineer, the
electronic architect, of the ERMETH (Elektronische Rechenmaschine der
ETH) which was built by him together with Stiefel and Rutishauser and
which was in use at ETH from 1956 to 1963.

Speiser was the founding director of the IBM Research Lab in
Rueschlikon, he was in charge of this lab for 10 years from 1956 to
1966. 1966 he became again the founding director of the new BBC/ABB
research lab in Baden-Daettwil. He was in this position from 1966 till
his retirement in 1987.

Speiser was very interested in the development and the activities of
ETH. He served for many years as a member of the board of ETH, he was
responsible for the political support that we needed to introduce a
curriculum for computer science at ETH in 1980/81. Speiser was
president of of the Swiss Academy of Engineering Sciences and also
president of IFIP already in the early years 1965 - 1968.

Those who attended the Latsis Symposium in February last year at ETH
Zurich, a conference commemorating 50 years of conjugate gradients and
celebrating Gene Golub's 70th birthday, might have met Ambros
Speiser. He was a great scientist and a true gentleman. We will miss him.

Walter Gander, ETH Zurich


From: Joanna Littleton <>
Date: Thu, 15 May 2003 15:17:12 -0400
Subject: Call for Nominations for Germund Dahlquist Prize

CALL FOR NOMINATIONS: The Germund Dahlquist Prize

The Germund Dahlquist Prize was established in 1995 and is awarded every
other year to a young scientist (normally under 45) for original
contributions to fields associated with Germund Dahlquist, especially the
solution of differential equations and numerical methods for scientific

The prize will be awarded at the SciCADE meeting in Trondheim, Norway, June
30 - July 4.


Nominations should include a letter of nomination, the candidate's CV, and
any other relevant information such as letters of support.

Nominations must be received in the SIAM office by JUNE 6, 2003. They
should be sent to:

Germund Dahlquist Prize Selection Committee
c/o Joanna Littleton
3600 University City Science Center
Philadelphia, PA 19104-2688

Selection Committee

The Prize Selection Committee is chaired by Linda R. Petzold (University of
California, Santa Barbara) and includes Christian Lubich (University of
Tuebingen, Germany), Gustaf Soderlind (Lund University, Sweden), and Andrew
Stuart (University of Warwick, UK).


From: Arieh Iserles <>
Date: Thu, 15 May 2003 09:48:12 +0100
Subject: Leslie Fox Prize Competition

The 11th LESLIE FOX PRIZE competition will take place on Friday, 20th
June, at the Isaac Newton Institute, Centre for Mathematical Sciences,
University of Cambridge.

Briefly, for those of you unfamiliar with the LFP, this is a biennial
prize given to the best young (under-31) numerical analysts, in two
stages: candidates submit essays, the best of which are shortlisted. The
shortlisted individuals all present 40 mins talks on the day. The
Adjudicators (this time: Arieh Iserles, Alastair Spence and Endre Suli)
choose the best talk(s) for the First Prize, other quality talks
receive Second Prizes.

The speakers this year are

* Melvin LEOK (California Institute of Technology)
* Adam OBERMAN (University of Texas)
* Marc SCHWEITZER (Universitaet Bonn)
* Tatjana STYKEL (University of Calgary)
* Jared TANNER (University of California at Davis)
* Boris VEXLER (Universitaet Heidelberg)

The subjects of their talks span the width and breadth of numerical
analysis and I can promise you a very exciting day! The past list of
LFP winners reads like a roll-call of young and up-coming numerical
analysts, most of whom have reached world prominence.

Attendance is free and everybody is welcome. There will be a lunch at
Wolfson Court and a dinner at Selwyn College (after-dinner speaker:
Nick Trefethen) which, needless to say, are not free. If you wish to
take part in either (or both) or need help with accommodation in
Cambridge, you need to register, but we welcome registration also
from casual participants.

One special incentive to come to this year's LFP is that it will be
held at the brand-new Centre for Mathematical Sciences, which is a
place worth visiting to all those interested in either mathematics or
architecture -- a guided tour will be arranged! Moreover, the one-day
meeting has been arranged in convenient time/space proximity to a
number of major conferences that will be held in Britain and Northern
Europe: Dublin differential equations conference, MAFELAP, the
biennial Dundee conference, and SciCADE.

More details and registration are available at the URL


From: Joseph Traub <>
Date: Mon, 12 May 2003 11:20:06 -0400 (EDT)
Subject: Announcement of a New Prize


This new annual award is for significant contributions to information-based
complexity by a young researcher. The prize will consist of $1000 and a plaque
and will be awarded at a suitable location.

Any researcher who has not reached their 35th birthday by September 30th
the year of the award is eligible.

The members of the Award Committee would appreciate nominations. These can
be sent to Joseph F. Traub via email at However, a
person does not have to be nominated to win the award.

The deadline for nominations for the first award is September 30, 2003.
The award can be for work done in a single year or a number of years. The
work can be published in any journal, number of journals or monographs.


From: Nick Sahinidis <>
Date: Fri, 16 May 2003 10:49:56 -0500
Subject: Sparse Boolean System

I wonder whether there exist any modulo 2 implementations of Gaussian
elimination for sparse matrices.

We are interested in identifying a non-zero solution of an under-determined
system of linear equations Ax=y, where A is m by n, m>>n, very sparse in
terms of 0-1 elements, and we are looking for 0-1 solutions for the vectors
x and y. I understand that solving diophantine equations is hard in
general but our matrix seems to have special structure that makes it
possible to solve this system via Gaussian elimination.

Any comments or pointers to the literature of related algorithms and
implementations would be appreciated.

Nick Sahinidis


From: W. B. Liu <>
Date: Mon, 12 May 2003 07:46:55 +0100 (BST)
Subject: Availability of Adaptive Finite Element Library

We are delighted to announce the availability of the finite element
library AFEPack 1.6 to the numerical analysis community.

AFEPack is a generic C++ adaptive finite element library. Comparing
with the existing finite element libraries, its contribution is twofold:
First, it introduces a framework of multi-meshes. That means you
can approximate multi-variables using very different adaptive
meshes, and this could save significant computational work. This feature
is particularly important when these variables have very different
computational difficulties.
Second, you can easily build very flexible finite element spaces with

The package has been tested on elliptic problems, parabolic problems,
flow problems, and several optimal control problems. Detailed documents
are included with the package.

AFEPack 1.6 now can be downloaded from
or obtained by sending an e-mail to
The document now can browsing at

Dr. R Li
School of Mathematics
Peking University, China

Prof. WB Liu
University of Kent
Canterbury, CT2 7PE UK


From: Jean Brac <>
Date: Wed, 14 May 2003 08:40:27 +0200
Subject: New Book on Wave Propagation

I could inform all members that a new book about wave propagation
is right now edited. This book is presently written in french
and a later version in english is planed by Springer Verlag.
That is why I provide you an english summary.
Thank you for you help to broadcast this news.
Jean Brac
English summary :
This book begins by the brief story of the wave concept through the
centuries but its main subject is about discrete scheme for acoustic and
elastic wave propagation.
It proposes a method of building numerical schemes of very high accuracy
based on the spectral analysis of the error. They are applied to wave
propagation but it can be used for any partial derivatives problem
solved by the means of finite difference.
Several formulations of the continue problem are presented but we keep
on the formulation in displacement velocities and strains. Moreover, an
analysis of the characteristics of the wave propagation equations leads
to compare them to the fluid mecanics equations and to show the
relationship conditions.
For discretization, Taylor's expansions at high order and an analysis of
the discretization error by Fourier transform are managed. Then, the
concept of optimal approximation is introduced in contrast with the the
approximations based on the truncature error. However the so-built
schemes remains convolutive. The efficiency of computation is huge in
front of the relative discretization error. A Fortran program provides
the calculation of the optimal coefficiens.
The time order elevation consists of translating the time high order
derivatives in space high order derivative. Finally, an stability
analysis is driven.
title : Propagation d'ondes acoustiques et


From: Andy Deelen <>
Date: Wed, 14 May 2003 08:03:20 +0100
Subject: New Book on Numerical Analsis of Wavelet Methods

Just published:

Numerical Analsyis of Wavelet Methods, by Albert Cohen, Studies in
Mathematics and its Applications, Volume 32, Hardbound, 356 pages, 2003,
356 pages, EUR/USD 95

SPECIAL OFFER PRICE: US$ / EUR 66 (please contact Andy Deelen at Elsevier
( for a special order form. Offer valid until 31
August 2003

This book is an updated and expanded version of a chapter published before
in the Handbook of Numerical Analysis, Volume VII; ISBN: 0-444-50350-1,


Since their introduction in the 1980's, wavelets have become a powerful tool
in mathematical analysis, with applications such as image compression,
statistical estimation and numerical simulation of partial differential
equations. One of their main attractive features is the ability to
accurately represent fairly general functions with a small number of
adaptively chosen wavelet coefficients, as well as to characterize the
smoothness of such functions from the numerical behaviour of these
coefficients. The theoretical pillar that underlies such properties involves
approximation theory and function spaces, and plays a pivotal role in the
analysis of wavelet-based numerical methods. This book offers a
self-contained treatment of wavelets, which includes this theoretical pillar
and it applications to the numerical treatment of partial differential
equations. Its key features are:

1. Self-contained introduction to wavelet bases and related numerical
algorithms, from the simplest examples to the most numerically useful
general constructions.

2. Full treatment of the theoretical foundations that are crucial for the
analysis of wavelets and other related multiscale methods : function spaces,
linear and nonlinear approximation, interpolation theory.

3. Applications of these concepts to the numerical treatment of partial
differential equations : multilevel preconditioning, sparse approximations
of differential and integral operators, adaptive discretization strategies.


Academic researchers in applied mathematics (in particular: numerical
analysis, partial differential equations, approximation theory, real
analysis). Engineers and academic researchers making use of numerical
simulation or image processing.

For furher information regarding this title, please see:

With kind regards,
Andy Deelen
Administrative Editor
Mathematics & Computer Science
Elsevier BV
Sara Burgerhartstraat 25
1055 KV Amsterdam
The Netherlands
tel: +31 20 485 2343
fax: +31 20 485 2616


From: Sara Ackerman <>
Date: Wed, 14 May 2003 11:13:56 -0400
Subject: New Book on Solving ODEs with MATLAB

New from Cambridge University Press...

"Solving ODEs with MATLAB" by L.F. Shampine, I. Gladwell and S. Thompson

This book is a text for a one-semester course for upper-level undergraduates
and beginning graduate students in engineering, science, and mathematics.
Prerequisites are a first course in the theory of ODEs and a survey course
in numerical analysis, in addition to specific programming experience,
preferably in MATLAB, and knowledge of elementary matrix theory.
Professionals will also find that this useful concise reference contains
reviews of technical issues and realistic and detailed examples. The
programs for the examples are supplied on the accompanying web site and can
serve as templates for solving other problems.

Table of Contents: 1. Getting Started 2. Initial Value Problems 3. Boundary
Value Problems 4. Delay Differential Equations

2003 / 272 pp. / 35 line diagrams
0-521-53094-6 Paperback: $35.00

Learn more about this exciting new text, including how to order it, by
visiting our online computer science catalog at:

Under the heading: Scientific Computing & Applications


From: Avraham Sidi <asidi@cs.Technion.AC.IL>
Date: Thu, 15 May 2003 14:24:36 +0300 (IDT)
Subject: New Book on Extrapolation Methods

Subject : New book on extrapolation (convergence acceleration)
Title : Practical Extrapolation Methods: Theory and Applications
Author : Avram Sidi
Publisher: Cambridge University Press
Series : Cambridge Monographs on Applied and Computational Mathematics
Number : 10
In print | Hardback | 542 pages | ISBN: 0-521-66159-5

The pdf file in
gives the complete table of contents and part of Chapter 1.

This state-of-the-art book focuses on the most effective nonlinear
extrapolation methods (or convergence acceleration methods,
or sequence transformations) that currently exist
in the literature. Extrapolation methods are used in the
fast and economical computation of limits (or antilimits) of
slowly convergent (or divergent) infinite sequences that arise
frequently in many areas of scientific computing and in different
engineering and scientific disciplines.

The methods considered in the book--approximately twenty in
number--are divided into two major categories:
(i) the Richardson extrapolation process and its generalizations,
(ii) sequence transformations.
The book gives an in-depth unified and up-to-date treatment of the
various methods that includes derivation, design of efficient algorithms,
and convergence and stability analyses, and provides many illustrative
examples. It shows how the methods studied can be tuned to achieve
numerical stability and high accuracy in practical computations.

This book is intended to serve both as a research monograph and
as a textbook, and should be of interest to researchers, graduate
students, and practitioners in the areas of applied mathematics,
numerical analysis, engineering, and theoretical physics and

Avram Sidi
Computer Science Department
Technion--Israel Institute of Technology


From: David F Griffiths <>
Date: Fri, 16 May 2003 15:47:23 +0100
Subject: Conference in Dundee on Numerical Analysis

Tuesday 24 June - Friday 27 June, 2003


The conference will be held at the West Park Conference Centre,
University of Dundee. The deadlines are
Submission of Abstracts: 23 May
Registration: 6 June

A limited number of short papers will be accepted for presentation (15
minutes and 5 minutes for discussion). We welcome papers on all
aspect of Numerical Analysis. Please submit a title and short
abstract via the web page.

Details of registration/accommodation fees and facilities for abstract
submission are available at the conference web site:

Conference Secretaries:
David Griffiths
Alistair Watson

Contact (email preferred)

Dr David F. Griffiths
Numerical Analysis Conference
Department of Mathematics
The University of Dundee
Dundee DD1 4HN
Scotland, UK

Telephone: +44(1382)344467/344471
FAX : +44(1382)345516


From: Laurence Yang <>
Date: Mon, 12 May 2003 07:15:16 -0300
Subject: Workshop in New Orleans on High Performance Computing


The 2nd Workshop on Hardware/Software Support for High Performance Scientific
and Engineering Computing (SHPSEC-03)
New Orleans, Louisiana, Sep. 27-- Oct. 1, 2003

The purpose of this workshop is to provide an open forum for computer
scientists, computational scientists and engineers, applied mathematicians,
and researchers to present, discuss, and exchange research-related ideas,
results, works-in-progress, and experiences in the areas of architectural,
compilation, and language support for problems in science and engineering

Possible topics include, but are not limited to:

1. Parallel and distributed architectures and computation models
2. Hardware/software co-design Compilers, hardware, tools, programming
languages and algorithms, operation system support and I/O issues
3. Techniques in the context of JAVA (Java multi-threading, Java processors
with parallelism, novel compiler optimizations for Java, etc.)
4. Compiler/hardware support for efficient memory systems
5. Novel uses for threads to improve performance or power
6. Software dynamic translation and modification
7. Hardware/software and technological support for energy-aware applications
8. Network, Mobile/wireless processing and computing
9. Software processes for development/maintenance/evolution of scientific and
engineering software
10.Cluster and grid computing

Further information about the conference proceedings and registration fee can
be found by web sites:

Important Deadlines:

Paper submission Due July 01, 2003
Notification of Acceptance August 01, 2003
Final camera-ready paper August 15, 2003

Workshop Organizers:

Prof. Minyi Guo
Dept. of Computer Software,
The University of Aizu
Aizu-Wakamatsu City, Fukushima, Japan

Prof. Laurence Yang
Department of Computer Science
St. Francis Xavier University
Antigonish, B2G 2W5, NS, Canada

Technical Committee: (see conference web page)


From: Simeon Simoff <>
Date: Sun, 18 May 2003 10:22:41 +1000
Subject: Workshop in Washington on Multimedia Data Mining


MDM/KDD2003 : The Fourth International Workshop on Multimedia Data Mining

in conjunction with

KDD-2003: 9th ACM SIGKDD International Conference on Knowledge Discovery &
Data Mining August 24 - 27, 2003, Washington, DC, USA

MDM/KDD2003 WORKSHOP THEME: Integrated Media Mining

Paper submission due 31 May 2003.

Workshop Website:
Workshop contact:
Workshop Date: 27 August, 2003


Valery A. Petrushin - Accenture Technology Labs, USA

Anne Kao - The Boeing Company, USA


Mihael Ankerst - The Boeing Company, USA
Chabane Djeraba - Nantes University, France
Marko Grobelnik - J. Stefan Institute, Slovenija
Latifur Khan - University of Texas at Dallas, USA
Simeon J. Simoff - University of Technology Sydney, Australia
Rod Tjoelker - The Boeing Company, USA


This year the Multimedia Data Mining workshop will bring together a diverse
group of academics and industry practitioners in integrated state-of-art
analysis of digital media content, multimedia database systems and
multimedia data streams. The workshop will address issues specifically
related to mining information from multi-modality, multi-source,
multi-format data in an integrated way. Many analysis domains collect data
from several sources, including static databases, streaming data, web pages,
or conditionally collected data. Data appear in multiple forms, including
structured, numeric, free text, video, image, speech, or combinations of
several types. Analysis in these domains requires combining of techniques
and integrating methods.

For details, please, visit the workshop Web site at:


From: Hans Bruun Nielsen <>
Date: Thu, 15 May 2003 15:36:30 +0200
Subject: Symposium in Copenhagen on Mathematical Programming


This is a reminder that the deadline for submission of abstracts for ISMP 2003
is May 31. It is possible to follow the registrations and the abstracts for
ISMP 2003 on-line:

Looking forward to seeing you in Copenhagen,

Jens Clausen and J=F8rgen Tind

The Mathematical Programming Society, the Technical University of
Denmark, and the University of Copenhagen announce:

ISMP 2003
Copenhagen, Denmark
August 18 - 22, 2003

The 18th International Symposium on Mathematical Programming takes
place August 18-22, 2003 at the Technical University of Denmark in
cooperation with University of Copenhagen. It is the main scientific
event of the Mathematical Programming Society held every 3 years on
behalf of the Society. The Symposium attracts more than one thousand
researchers from all areas of mathematical programming. At the
symposium homepage you find all information including
the list of plenary and semiplenary speakers.

A proceedings volume sponsored by Springer Verlag with a contribution
from each speaker will be published as a special issue of Mathematical
Programming and will be distributed to all participants at the

The scientific program will be complemented by an attractive social
program: At ISMP 2003 you will have the opportunity to taste the
traditional "City Hall" pancakes served at the City Hall and to
participate in a cocktail party in the Celebration Hall at the
University of Copenhagen. The conference dinner will take place at the
famous Base Camp restaurant in the former naval area of the city,
recently rebuilt into one of the main cultural centers of the city.
Additional activities will be arranged in the scenic surroundings of
the Campus.

The homepage provides information about registration, travel, and
accomodation, and offers possibilities for web-based abstract
submission of contributed presentations as well as organized sessions.

Program committee:

J=F8rgen Tind (chair), University of Copenhagen
Jens Clausen (cochair), Technical University of Denmark
Rainer Burkard, Technische Universit=E4t, Graz
Martin Gr=F6tschel, ZIB, Berlin
Michael J. Todd, Cornell University
Stephen J. Wright, University of Wisconsin

Local organizing committee:

Jens Clausen (chair), Technical University of Denmark
J=F8rgen Tind (cochair), University of Copenhagen
Hans Bruun Nielsen, Technical University of Denmark
David Pisinger, University of Copenhagen
Martin Zachariasen, University of Copenhagen

Important dates:

May 31, 2003: Deadline for submission of abstracts.
August 18 - 22: ISMP 2003

We look forward to welcoming you in Copenhagen.

J=F8rgen Tind and Jens Clausen


From: William Dai <>
Date: Thu, 15 May 2003 13:29:24 -0600
Subject: Staff Position at Los Alamos National Lab

The High Performance Computing Environments Group of Los Alamos National
Lab currently has openings for staff member positions. We are looking for
candidates who have background in physics, or mathematics, or computer
science, and have experience in parallel computing. Any experience in
development of numerical algorithm for partial differential equations for
physics/engineering problems, eg., fluid dynamics, is valuable in our

Potential candidates may send resume to

William Dai


From: Bart Truyen <>
Date: Sat, 17 May 2003 16:12:36 +0200
Subject: Ph.D. Studentship at Vrije Universiteit Brussel

Applications are invited for a Ph.D. studentship in the areas of numerical
analysis and inverse problems. The candidate is expected to conduct research
on the numerical aspects of Electrical Impedance Tomography (EIT, also known
as resistance imaging).=A0 Particular emphasis will be on the development and
application of subspace methods for the solution of structured nonlinear
problem formulations. The remarkable stability properties of these methods
will be further studied, based on the concept of principle angles.
Additional information about the subject and related research projects can
be found in the accompanying web pages,

Appointments will be for 4 years, with a two-year probationary period.
Candidates must have completed a M.Eng., M.Sc., or equivalent degree in
applied mathematics, electrical engineering, or physics. Special
consideration will be given to candidates with a demonstrated experience in
numerical linear algebra, inverse problems, or a closely related subject in
the field of tomographic imaging. The research position will entail the
preparation of a Ph.D. degree in Engineering or Applied Sciences (possibly
in collaboration with the originating research institute).

Successful applicants will find at the Department of Electronics and
Information Processing ETRO, a stimulating research environment conducive to
professional growth, providing various collaborative research opportunities
and the possibility of travel.

Inquiries and applications should be addressed to Bart Truyen, Department of
Electronics and Information Processing (ETRO), Vrije Universiteit Brussel -
VUB, Pleinlaan 2, B-1050, Brussels, Belgium,,
The Vrije Universiteit Brussel - VUB is an equal opportunity/affirmative
action employer.


End of NA Digest