- Today's Editor:
- Cleve Moler
- The MathWorks, Inc.
- moler@mathworks.com

- Linear Algebra Problem
- Numerical Methods Applied to Agriculture
- Mailing List Devoted to Teaching Complex Analysis
- Computational Aeroacoustics
- Birthday of Modern Numerical Analysis
- New Book on Air Pollution Dispersion Modeling
- New Book, Nonconvex Optimization in Mechanics
- Workshop on Applied Parallel Computing
- Summer School in Numerical Analysis
- European Symposium on Algorithms
- ICFD Conference on Numerical Methods for Fluid Dynamics
- Benelux Signal Processing Symposium
- BLAS Technical Forum
- Workshop on Numerical Ordinary Differential Equations
- Faculty Positions at Worcester Polytechnic Institute
- Faculty Position at Temple University
- Postdoctoral Position at Manchester
- Postdoctoral Position at Sandia National Laboratories
- Research Position at University of Liverpool
- Research Positions at Athens University
- Student Positions at University of Texas at San Antonio
- Contents, Linear Algebra and Its Applications
- Contents, BIT
- Contents, Mathematics of Control, Signals, and Systems

**URL for the World Wide Web:**
http://www.netlib.org/na-net/na_home.html

-------------------------------------------------------

From: Erich Novak <novak@mi.uni-erlangen.de>

Date: Tue, 11 Nov 1997 22:35:00 GMT

**Subject: Linear Algebra Problem**

Dear colleague,

I have the following PROBLEM IN LINEAR ALGEBRA, I do not know the answer.

Assume that d and n are natural numbers and define $f: \R^d \to \R$ by

$$

f(x) = (\prod_{l=1}^d \cos^2(x^l)) - 1/n ,

$$

where $x=(x^1,...,x^d)$. Hence $x^l$ is the $l$th component

of the vector $x$. Prove or disprove the following CONJECTURE:

For any given $x_1,...,x_n \in \R^d$ the $(n,n)$-matrix

$A$ given by

$$

a_{ij} = f(x_i-x_j)

$$

is positive semidefinite, i.e., the eigenvalues are nonnegative.

(Comment: I know that this is true for $n \ge 2^d$. So the

interesting case would be $n < 2^d$.)

Erich Novak novak@mi.uni-erlangen.de

------------------------------

From: Paulo Fernando de Arruda Mancera <Caar@botunet.com.br>

Date: 11 Nov 1997 21:42:02 EDT

**Subject: Numerical Methods Applied to Agriculture**

Does anyone know about numerical methods which are applied

to solve agriculture and irrigation problems?

Thanks.

Paulo

Paulo Fernando de Arruda Mancera

Departamento de Bioestatistica

IB/UNESP

Rubiao Jr

Botucatu

18618-000

Brazil

e-mail: caar@botunet.com.br

------------------------------

From: Ron Buckmire <ron@abacus.oxy.edu>

Date: Mon, 10 Nov 1997 18:23:26 -0800 (PST)

**Subject: Mailing List Devoted to Teaching Complex Analysis**

CA-TEACH

11/15/97

CA-TEACH is an internet mailing list devoted to the discussion of the teaching

of COMPLEX ANALYSIS. Though it is intended to be a list devoted to teaching

in an undergraduate setting, it is not restricted to this.

Regular topics on CA-TEACH could include: a discussion of Complex Analysis

textbooks, broadcasts of URLs where documents on complex analysis can be

found, personal reports on teaching complex analysis courses at your

institution, impact of technology on the teaching of complex analysis,

curriculum debates, lecture notes and examples of using various software

packages to assist in complex analysis visualization, among others.

CA-TEACH will *not* be a moderated list. If people are interested in a

moderated list then I will gladly accept volunteers to be the moderator.

To join the CA-TEACH mailing list, send a message to MAJORDOMO@abacus.oxy.edu

with one line in the message body

SUBSCRIBE CA-TEACH

Any other questions about the list should be sent to

CA-TEACH-approval@abacus.oxy.edu or Majordomo-Owner@abacus.oxy.edu

RON BUCKMIRE

Math Dept., Occidental College

------------------------------

From: C. H. Lai <C.H.Lai@greenwich.ac.uk>

Date: Wed, 12 Nov 1997 09:42:26 GMT

**Subject: Computational Aeroacoustics**

We would like to announce the web page on our work on Computational

Aeroacoustics

http://www.gre.ac.uk/~dg06

Computational Aeroacoustics Group

School of Computing and Mathematical Sciences

University of Greenwich

------------------------------

From: Joe Grcar <sepp@california.sandia.gov>

Date: Mon, 10 Nov 97 19:37:14 -0800

**Subject: Birthday of Modern Numerical Analysis**

It seems this anniversary will not be celebrated in conferences and

journals, so before it passes I am commemorating the:

50th Birthday of Modern Numerical Analysis

November 1947 -- November 1997

When computers were invented, John von Neumann saw the accuracy

of calculations would need study. He made an example of Gaussian

elimination because many people thought it would be unsuitable for

automatic computing. At Princeton, in the winter of 1946-47, von

Neumann proved the computer he was building could invert matrices

larger than thought possible at the time. The proof appeared in

"Numerical Inverting of Matrices of High Order" [Bull AMS, Nov '47]

written with Herman Goldstine. Few people read the paper in detail,

but some who mattered did. Wallace Givens and James Wilkinson

adopted the paper's approach, and made the paper the model for

rounding error analysis (alternatives now include automatic,

interval and statistical analyses).

Although many parts of numerical analysis existed before von

Neumann's paper, they coalesced as an academic discipline in the

subsequent decade. Von Neumann and Goldstine's paper has been

called the first in this "modern" numerical analysis because it is

the first to study rounding error and because much of the paper is

an essay on scientific computing (albeit with an emphasis on

numerical linear algebra). The list of error sources in Chapter 1

is clearer and more authoritative than any since. The axiomatic

treatment of rounding error in Chapter 2 inspired later analyses.

The discussion of linear algebra in Chapters 3 to 5 contains

original material, including the invention of triangular factoring.

(Von Neumann is the last of three inventors. The LDU factorization

in particular should be named for him.)

The rounding error analysis in Chapter 6 accounts for just

one-quarter of the paper, with the analysis of triangular factoring

a fraction of that. Von Neumann shows, for symmetric positive

definite matrices, the backward error of factoring equals the sum of

the errors incurred on the Gaussian elimination steps. In the many

years following the paper, no other condensation of the errors has

been proposed. Von Neumann bounds the backward error using this

sum. The bulk of Chapter 6 then bounds the residual of inverting

symmetric positive definite matrices. General matrices are treated

by applying the preceding to something like the normal equations.

Textbooks do a great disservice when they note --- without

explaining why --- von Neumann proved backward error bounds for

factoring positive definite matrices. As a result, many people are

unaware this is von Neumann's discovery and not his oversight. It

is difficult to learn the truth because the matter is not discussed

plainly, but there seem to be only two predictive ("a priori" in the

subject's jargon) backward error bounds. For matrices of order n,

the bound for von Neumann's positive definite case varies by n,

while the bound for the general case varies by 2**n. Von Neumann

even remarked he discovered the positive definite case because he

could not find "satisfactory" error bounds for factoring general

matrices. Small, predictive bounds are the only proof an algorithm

is consistently accurate. To explain things plainly, after 50 years

of work, the only acceptable bound we have for the most widely used

algorithm is the bound we started with, von Neumann's.

The concluding Chapter 7 interprets the rounding error analysis.

Von Neumann asked his readers to continue from his residual bound

"several different ways". He guided his readers by explaining the

appropriate conclusion shows the computed result is exact for some

perturbation of the initial data. This is the "backward"

interpretation which many people think von Neumann did not

understand. The paper closes by evaluating the residual bound for

"random" matrices, and by counting arithmetic operations.

In sum, von Neumann's paper contains much that is unappreciated

or at least unattributed to him. The contents are so familiar, it

is easy to forget von Neumann is not repeating what everyone knows.

He anticipated many of the developments in the field he originated,

and his theorems on the accuracy of Gaussian elimination have not

been encompassed in half a century. The paper is among von

Neumann's many firsts in computer science. It is the first paper in

modern numerical analysis, and the most recent by a person of von

Neumann's genius.

Happy Birthday and Best Regards from Joe Grcar

ps 1. Klara von Neumann named a puppy after her husband's matrix

inversion project. John, Klara and a fully grown Inverse can be

seen in "Passing of a Great Mind", Life, Feb. 25, 1957. Also note

the photo of von Neumann by the famous Time-Life photographer Alfred

Eisenstaedt.

ps 2. Wilkinson's "von Neumann Lecture" [SIAM Rev, '71] takes

passages of von Neumann's paper out of context and recommends

obvious improvements. This has been interpreted to mean the paper

is somehow flawed. As Paul Halmos explained, von Neumann many times

gave lesser mathematicians the opportunity to "improve" von Neumann.

ps 3. Von Neumann and Goldstine's four errors of scientific

computing are these. Last is precision: no computing device can

perform all its operations exactly, and when an imperfect operation

is performed, it injects noise into the calculation. Third is

approximation: the formulas of scientific theories must be made

amenable to evaluation by machine operations, and unending searches

for results must be terminated. Second is observation: physical

data must be ascertained by measurement either directly or through

other calculations. First is theory: the problem underlying the

calculation may be idealized, but further, any mathematical

formulation of a physical problem "necessarily represents only a

theory of reality, and not reality itself."

ps 4. "We may interpret the elimination method as the

combination of two tricks", von Neumann remarked as he described a

complicated algorithm by a simple relation among matrices. Analysis

and design of algorithms (many even are called matrix algorithms

today) would be impossible without this kind of simplification.

Thus it seems likely the equivalence between Gaussian elimination

and triangular factoring is a paradigm for numerical analysts, in

the sense of Thomas Kuhn. Considering its importance, its history

is too much ignored. Triangular factoring apparently originated

with T. Banachiewicz [Bull Int L'Acad Polonaise, Ser A, '38], and

later independently with P. S. Dwyer [Ann Math Stat, '44], and then

von Neumann and Goldstine, [Bull AMS, '47]. Perhaps someone can

tell Banachiewicz' story.

------------------------------

From: Milton Beychok <mbeychok@deltanet.com>

Date: Thu, 13 Nov 1997 02:00:05 -0800

**Subject: New Book on Air Pollution Dispersion Modeling**

Anyone interested in air pollution dispersion modeling is invited to

visit this website:

http://www.air-dispersion.com

to learn about "FUNDAMENTALS OF STACK GAS DISPERSION", a comprehensive

single-source reference book on dispersion modeling of continuous,

buoyant pollution plumes. The website provides a brief description of

the book, peer reviews published in technical and scientific journals,

the book's complete table of contents, and information on how to obtain

copies.

Topics covered in the book include: classifying atmospheric stability;

determining of buoyant plume rise; Gaussian dispersion equations and

models; developing windspeed profiles; trapped pollution plumes;

fumigated plumes; flare stack plumes; meteorological parameters, and

much more.

Milton Beychok, 2233 Martin St. # 205, Irvine, CA 92612, USA

Phone & Fax: 714-833-8871

E-mail (1): mbeychok@air-dispersion.com

E-mail (2): mbeychok@deltanet.com

------------------------------

From: Georg Stavroulakis <gs@r2.infam.bau.tu-bs.de>

Date: Sat, 15 Nov 1997 18:36:52 +0100

**Subject: New Book, Nonconvex Optimization in Mechanics **

Announcement of New Book

NONCONVEX OPTIMIZATION IN MECHANICS.

Algorithms, Heuristics and Engineering Applications by the F.E.M.

by EURIPIDIS S. MISTAKIDIS

Institute of Steel Structures, Dept. of Civil Engineering,

Aristotle University, Thessaloniki, Greece

and GEORGIOS E. STAVROULAKIS

Institute of Applied Mechanics, Dept. of Civil Engineering,

Carolo Wilhelmina Technical University, Braunschweig, Germany

NONCONVEX OPTIMIZATION AND ITS APPLICATIONS Vol. 21

KLUWER ACADEMIC PUBLISHERS, ISBN 0-7923-4812-5, December 1997, 300 pp.

Price: NLG 260.00 / USD 149.00 / GBP 89.00

Available at a reduced price for course adoption

when ordering six copies or more.

For more details and online order, see

http://www.wkap.nl/book.htm/0-7923-4812-5

DESCRIPTION: This book presents, in a comprehensive way, the application

of optimization algorithms and heuristics in engineering problems

involving smooth and nonsmooth energy potentials. These problems arise in

real-life modeling of civil engineering and engineering mechanics

applications. Engineers will gain an insight into the theoretical

justification of their methods and will find numerous extensions of the

classical tools proposed for the treatment of novel applications with

significant practical importance. Applied mathematicians and software

developers will find a rigorous discussion of the links between applied

optimization and mechanics which will enhance the interdisciplinary

development of new methods and techniques. Among the large number of

concrete applications are unilateral frictionless, frictional or adhesive

contact problems, and problems involving complicated friction laws and

interface geometries which are treated by the application of fractal

geometry. Semi-rigid connections in civil engineering structures, a topic

recently introduced by design specification codes, complete analysis of

composites, and innovative topics on elastoplasticity, damage and optimal

design are also represented in detail.

AUDIENCE: The book will be of interest to researchers in mechanics, civil,

mechanical and aeronautical engineers, as well as applied

mathematicians. It is suitable for advanced undergraduate and graduate

courses in computational mechanics, focusing on nonlinear and nonsmooth

applications, and as a source of examples for courses in applied

optimization.

CONTENTS:

Part I: Nonconvexity in Engineering Applications.

1. Nonconvexity in Engineering Applications.

Part II: Applied Nonconvex Optimization Background.

2. Applied Nonconvex Optimization Background.

Part III: Superpotential Modelling and Optimization in Mechanics

with and without Convexity and Smoothness.

3. Convex Superpotential Problems.

4. Nonconvex Superpotential Problems.

5. Optimal Design Problems.

Part IV: Computational Mechanics. Computer Implementation,

Applications and Examples.

6. Computational mechanics algorithms

7. Applications.

Index.

------------------------------

From: Jerzy Wasniewski <unijw@unidhp1.uni-c.dk>

Date: Sun, 9 Nov 1997 21:17:48 +0100 (MET)

**Subject: Workshop on Applied Parallel Computing**

PARA98 - International Workshop on Applied Parallel Computing,

Large Scale Scientific and Industrial Problems,

June 14-17, 1998, Umea University, Sweden

(see http://www.hpc2n.umu.se/para98/)

Call for Papers and 2nd Announcement

High Performance Computing Center North (see http://www.hpc2n.umu.se)

is hosting the fourth International Workshop on Applied Parallel

Computing (PARA98) in June 14-17, 1998 at Umea University, Sweden.

The general theme for PARA98 is Large Scale Scientific and

Industrial Problems focusing on:

o High-performance computing applications in academia and industry,

o Tools, languages and environments for high-performance computing,

o Scientific visualization and virtual reality applications in

academia and industry,

o Future directions in high-performance computing and networking.

The PARA98 meeting is aimed to be an international forum for

idea and competence exchange for specialists in parallel computing,

visualization, etc and scientists from industry and academia

solving large scale computational problems.

Another important aim of the PARA meetings is to strengthen the ties

between HPC centers, academia, and industry in the Nordic countries

as well as worldwide.

The meeting starts with a one day tutorial followed by a three day

workshop. There will be several invited one-hour lectures as well as

contributed 20-30 minutes talks. The conference proceedings will be

published by Springer Verlag in their LNCS series.

Deadlines, Abstracts and Papers:

(see http://www.hpc2n.umu.se/para98/ for details)

o February 15, 1998: Extended Abstracts

o April 15, 1998: Notification to authors

o July 31, 1998: Papers (between 5 and 10 pages)

Questions regarding PARA98 should be addressed to para98@hpc2n.umu.se.

The PARA Steering Committee:

- Petter Bjorstad, University of Bergen (Norway)

- Jack Dongarra, University of Tennessee and Oak Ridge National Lab (USA)

- Bjorn Engquist, PDC, Royal Institute of Technology, Stockholm (Sweden)>

- Kristjan Jonasson, University of Iceland, Reykajavik (Iceland)

- Bo Kagstrom, Umea University and HPC2N (Sweden), PARA98 Chairman

- Risto Nieminen, Helsinki University of Technology, Espoo (Finland)

- Karstein Sorli, SINTEF, Dept of Industrial Mathematics, Trondheim (Norway)

- Olle Teleman, Center for Scientific Computing (CSC), Espoo (Finland)

- Jerzy Wasniewski, Danish Computing Centre for Research and Education

(UNI-C), Lyngby (Denmark), PARA94-96 Chairman

------------------------------

From: Gabriel Gatica <Gabriel.Gatica@ing-mat.udec.cl>

Date: Sun, 09 Nov 1997 17:55:21 -0800

**Subject: Summer School in Numerical Analysis**

FIRST SUMMER SCHOOL OF THE

F O N D A P I N A P P L I E D M A T H E M A T I C S:

Numerical Analysis and Mathematical Modelling

CONCEPCION, CHILE, January 19 - 30, 1998

http://www.ing-mat.udec.cl/fondap/escuela.html

The National Foundation for Scientific and Technological Research of

Chile (CONICYT-CHILE), through its Program "FONDO NACIONAL DE AREAS

PRIORITARIAS" (F O N D A P) in APPLIED MATHEMATICS, and the Department

of Mathematical Engineering of the University of Concepcion, are very glad

to invite you to attend the FIRST SUMMER SCHOOL ON NUMERICAL ANALYSIS AND

MATHEMATICAL MODELLING, to be held at the University of Concepcion,

Concepcion, Chile, during the period January 19 - 30, 1998.

The following courses will be taught:

1) BOUNDARY ELEMENTS, THE h-p VERSION.

Prof. Ernst P. Stephan, University of Hannover, Germany.

2) INTRODUCTION TO FINITE VOLUME METHODS FOR HYPERBOLIC

SYSTEMS OF CONSERVATION LAWS.

Prof. Frederic Coquel, University of Paris VI, France.

3) MATHEMATICAL MODELLING OF SOME PROBLEMS

IN ENGINEERING.

Prof. Alfredo Bermudez C., University of Santiago of Compostela, Spain.

4) VARIATIONAL METHODS OF BOUNDARY INTEGRAL EQUATIONS:

THEORY AND APPLICATIONS.

Prof. George C. Hsiao, University of Delaware, USA.

5) WAVELET METHODS IN NUMERICAL SIMULATION.

Prof. Albert Cohen, University of Paris VI, France.

F E L L O W S H I P S

There are a limited number of fellowships (about 50) for financing

registration fees, round trip tickets and stay expenses of chilean

and foreigners students (only graduate and advanced undergraduate

students)

Secretaria de la Escuela de Verano FONDAP

Departamento de Ingenieria Matematica

Facultad de Ciencias Fisicas y Matematicas

Universidad de Concepcion

Casilla 4009, Concepcion, Chile

Fax: 056-41-251529, Tel.: 056-41-203148/203118/

/203121/204539

esc.verano98@ing-mat.udec.cl

http://www.ing-mat.udec.cl/fondap/escuela.html

------------------------------

From: Giuseppe Italiano <italiano@dsi.unive.it>

Date: Mon, 10 Nov 1997 10:51:48 +0100 (MET)

**Subject: European Symposium on Algorithms**

Preliminary Call For Papers

Sixth Annual European Symposium on Algorithms

ESA'98

August 24--26, 1998, Venice, Italy

The 6th Annual European Symposium on Algorithms (ESA '98) will be held

in Venice, Italy, August 24--26, 1998. The Symposium covers research

in the use, design, and analysis of efficient algorithms and data

structures as it is carried out in computer science, discrete applied

mathematics and mathematical programming. Papers are solicited

describing original results in all areas of algorithmic research,

including but not limited to: Approximation Algorithms; Combinatorial

Optimization; Computational Biology; Computational Geometry; Databases

and Information Retrieval; Graph and Network Algorithms; Machine

Learning; Number Theory and Computer Algebra; On-line Algorithms;

Pattern Matching and Data Compression; Symbolic Computation. The

algorithms may be sequential, distributed or parallel, and they should

be analyzed either mathematically or by rigorous computational

experiments. Submissions that report on experimental and applied

research are especially encouraged.

Further Information: All questions should be mailed to esa98@dsi.unive.it.

Additional information may be found through anonymous FTP or

World-Wide Web:

FTP: ftp.dsi.unive.it/priv/esa98/

WWW: http://www.dsi.unive.it/~esa98

------------------------------

From: Bette Byrne <Bette.Byrne@comlab.ox.ac.uk>

Date: Mon, 10 Nov 97 12:34:49

**Subject: ICFD Conference on Numerical Methods for Fluid Dynamics**

Information on the ICFD Conference on Numerical Methods for Fluid Dynamics

to be held at Oxford, 31 March - 3 April 1998 can now be found on the

following web site:

http://www.comlab.ox.ac.uk/oucl/icfd/1annb.htm

or

http://www.comlab.ox.ac.uk/oucl/icfd under "international conference"

Bette Byrne

ICFD Secretary

Oxford University Computing Laboratory

------------------------------

From: Marc Moonen <Marc.Moonen@esat.kuleuven.ac.be>

Date: Thu, 13 Nov 1997 16:41:00 +0100

**Subject: Benelux Signal Processing Symposium**

IEEE BENELUX SIGNAL PROCESSING CHAPTER

SIGNAL PROCESSING SYMPOSIUM

March 26-27, 1998, Leuven, Belgium

2nd ANNOUNCEMENT AND CALL FOR PAPERS

The first IEEE Benelux Signal Processing Symposium, sponsored by the

IEEE Benelux Signal Processing Chapter, will be held March 26-27, 1998

in the K.U.Leuven Faculty Club, Leuven, Belgium.

The program will consist of invited tutorial and selected oral

presentations, as well as informal poster sessions.

Invited presentations :

Pierre Duhamel (ENST Paris, France)

Signal processing in source compression : is PR compulsory ?

Can we reduce artifacts due to DCT coding based on DCT properties?

Peter Grant (University of Edinburgh, U.K.)

*IEEE Signal Processing Society Distinguished Lecturer*

CDMA array processing

Inald Lagendijk (TU Delft, The Netherlands)

Watermarking signal processing techniques for video copy protection

Steve Young (Cambridge University, U.K.)

Large vocabulary continuous speech recognition

Papers are solicited for technical sessions on the following and

related topics :

Digital Signal Processing

Statistical Signal and Array Processing

Speech processing

Audio and Electroacoustics

Image and Multidimensional Signal Processing

Knowledge Engineering and Signal Processing

Implementations

Applications

Workshop Secretariat

Mrs. Ida Tassens

Dept. of Electrical Engineering

Kath. Universiteit Leuven

K. Mercierlaan 94

B-3001 Heverlee, Belgium

tel : +32-16-32 17 09,

fax : +32-16-32 19 70

Ida.Tassens@esat.kuleuven.ac.be

Chairmen

Marc Moonen

Kath. Universiteit Leuven

Piet Sommen

Technische Universiteit Eindhoven

------------------------------

From: Jack Dongarra <dongarra@cs.utk.edu>

Date: Thu, 13 Nov 1997 21:23:55 -0500

**Subject: BLAS Technical Forum**

BLAS TECHNICAL FORUM, December 3-5, 1997, Knoxville, TN

We are planning a BLAS Technical Forum meeting on December 3-5.

The meeting will be hosted by the University of Tennessee and

will be held at the Hilton Hotel in Knoxville, TN:

Hilton Hotel

501 W Church Street

Knoxville, TN

(423) 523-2300

(800) 445-8667

The Forum has been established to consider expanding the Basic Linear

Algebra Subprograms (BLAS) in a number of directions in the light of

modern software, language, and hardware developments. The first meeting

of the Forum was held in Nashville on February 19-20, 1996 and the

previous meeting was in Portland, OR on August 14-15, 1997.

Working groups have been established to consider the overall

functionality, possible extensions, and a lightweight interface

for the BLAS, as well as the short term goals of the forum. Other

subgroups have also been established, either to advise the current

working groups or as placeholders for future working groups on parallel

processing issues, sparse operations, and language binding issues.

We strongly urge and encourage attendance at the meeting

so that we can make tangible progress towards much needed standards.

Wide input is needed to help ensure that emerging proposals are useful

and acceptable to the community.

It is appreciated that it is not easy for everyone to attend the

meetings of the Forum, but we would nevertheless welcome your input

since we wish the discussion to be as open as possible, and the results

to reflect consensus from the community at large.

We will plan to start the meeting with lunch at 12:00 pm on Wednesday,

December 3rd and end by Noon on Friday, December 5th. There will

be no registration fee, and lunch is provided on December 3 and 4.

Refer to the BLAS Technical Forum webpage for complete details.

http://www.netlib.org/utk/papers/blast-forum.html

Jack Dongarra and Sven Hammarling

------------------------------

From: John Butcher <butcher@math.auckland.ac.nz>

Date: Fri, 14 Nov 1997 16:18:26 +1300

**Subject: Workshop on Numerical Ordinary Differential Equations**

Workshop on

Numerical ordinary differential

equations in theory and practice

29 June 1998 - 10 July 1998

Auckland, New Zealand

This workshop will feature two main series of lectures by prominent

numerical analysts. There will also be scheduled lectures by other

participants. Provision will be made for impromptu presentations on

work, as it develops, arising from discussions amongst people taking

part in the workshop. The venue will be the City Campus of the

University of Auckland.

Invited speakers

The following internationally known speakers have agreed to present

series of lectures at the workshop

Professor J. M. ("Chus") Sanz-Serna, Valladolid, Spain

Professor Marc Spijker, Leiden, The Netherlands

Expressions of interest

To express interest in taking part in this workshop and to have your

name added to the mailing list, please e-mail the address

ANODE@math.auckland.ac.nz

or a member of the organising committee.

Organising committee

John Butcher butcher@math.auckland.ac.nz

Robert Chan chan@math.auckland.ac.nz

Allison Heard heard@math.auckland.ac.nz

Further information

For further information please visit our website

http://www.math.auckland.ac.nz/~anode/

------------------------------

From: Homer Walker <walker@WPI.EDU>

Date: Tue, 11 Nov 1997 20:05:34 -0500 (EST)

**Subject: Faculty Positions at Worcester Polytechnic Institute**

POSITIONS IN APPLIED AND COMPUTATIONAL MATHEMATICS

DEPARTMENT OF MATHEMATICAL SCIENCES

WORCESTER POLYTECHNIC INSTITUTE

The Mathematical Sciences Department of Worcester Polytechnic Institute

(WPI) invites applications for anticipated tenure-track faculty positions

in applied and computational mathematics beginning in the fall of 1998.

Appointments will probably be at the Assistant Professor level, but

exceptionally well qualified candidates may be considered for appointments

at higher rank. An earned Ph.D. or equivalent degree is required. A

successful candidate must be able to contribute strongly to both the

department's research activities and its innovative, project-based

educational programs. Areas of research in the department include partial

differential equations with applications in fluid and continuum mechanics,

composite materials, computational modeling and simulation, numerical

linear algebra and nonlinear equations, optimization, control theory,

applied probability, discrete mathematics, and applied statistics.

WPI is an innovative technological university of engineering, science,

management, and the humanities and arts. It is private and highly selective,

with an enrollment of 2700 undergraduates and about 1000 full- and part-time

graduate students, and is ranked among the top 50 national universities.

The WPI campus is located in Worcester, MA, New England's second largest

city, in close proximity to the city's many cultural attractions as well

as nine other institutions of higher education in the urban area. Worcester,

forty miles west of Boston, offers access to the diverse cultural and

recreational resources of New England and provides opportunities for

urban, suburban or rural lifestyles. WPI offers a smoke-free environment.

The Mathematical Sciences Department has 23 full-time faculty and

supports a PhD program and MS programs in applied mathematics and

applied statistics, as well as a full undergraduate program. For

additional information about the Mathematical Sciences Department

and WPI, see http://www.wpi.edu/Academics/Depts/Math/.

Qualified applicants should send a detailed curriculum vitae, a one-page

statement of their specific teaching and research objectives, and the names

of four references with mail/email addresses and telephone/fax numbers to

Search Committee

Applied and Computational Mathematics

Mathematical Sciences Department

WPI

100 Institute Road

Worcester, MA 01609-2280, USA

Applicants will be considered on a continuing basis beginning January 1,

1998 until the position is filled.

To enrich education through diversity, WPI is an affirmative action, equal

opportunity employer.

------------------------------

From: Daniel Szyld <szyld@euclid.math.temple.edu>

Date: Fri, 14 Nov 1997 15:42:54 -0500 (EST)

**Subject: Faculty Position at Temple University**

TEMPLE UNIVERSITY

Department of Mathematics

The Department of Mathematics at Temple University invites

applications for a tenure-track position at the Assistant

Professor level, pending budgetary approval.

Applicants must have a Ph.D. and a solid record of

accomplishment or outstanding promise in research.

Priority will be given to applied mathematics, especially

scientific computing.

Temple has made a strong commitment to excellence in

undergraduate education. Candidates should have a proven record

of effectiveness in teaching mathematics at all levels.

The position will start in the fall of 1998. Salary is competitive.

Research achievements and teaching together with significant

contributions to our undergraduate programs will be the primary

criteria for advancement.

Women and minorities are particularly invited to apply.

We request that applicants send the AMS Application Cover Sheet

along with their vita and at least three letters of recommendation to:

Search Committee

Department of Mathematics

Temple University

Philadelphia, PA 19122

The vita should include a description of the applicant's teaching

experience and evidence of teaching ability, and a statement

of professional goals.

The application deadline is January 16, 1998.

For further information, see

http://www.math.temple.edu/position.fold/position.html

------------------------------

From: Nick Higham <higham@ma.man.ac.uk>

Date: Fri, 14 Nov 1997 13:39:25 GMT

**Subject: Postdoctoral Position at Manchester**

The Engineering and Physical Sciences Research Council are funding a

Research Assistantship for a project titled

``Numerical Analysis of the Generalized Eigenvalue Problem''.

This is a three-year post, starting in January 1998,

or as soon as possible thereafter.

The successful applicant will join the Numerical Analysis group at the

University of Manchester. The aim of the project is to develop theory,

algorithms and software for the generalized eigenvalue problem.

Applicants should hold, or expect to complete before the start date,

a Ph.D., and should have a strong background in numerical analysis,

linear algebra, and software development.

The appointment will be made at the RA 1A level, with a current

starting salary of 16,927 pounds sterling per annum.

Applications and requests for further information should be sent to

Professor N. J. Higham

Department of Mathematics, University of Manchester

Manchester, M13 9PL, England

Phone: 0161 275 5822

Fax: 0161 275 5819

email: higham@ma.man.ac.uk

http://www.ma.man.ac.uk/~higham/

Applicants should send a CV and arrange to have two letters of references

sent, all to the above address. Closing date: December 1, 1997.

------------------------------

From: Karen Devine <kddevin@quincey.cs.sandia.gov>

Date: Wed, 12 Nov 97 16:20:39 MST

**Subject: Postdoctoral Position at Sandia National Laboratories**

Post-Doctoral Position

Parallel Computing Sciences Department

Sandia National Laboratories

The Parallel Computing Sciences Department at Sandia National Laboratories

is seeking qualified candidates for a Post-Doctoral position to contribute to

the development of dynamic load-balancing libraries for scientific simulations

on parallel computers. Candidates must be U.S. citizens, have earned a

Ph.D. or equivalent, and have experience with the design and implementation

of dynamic load-balancing algorithms on distributed memory computers.

Experience with algorithm development on heterogeneous parallel systems

is particularly desirable. Candidates should be familiar with numerical

techniques such as adaptive finite element and difference methods. Strong

communication skills will also be needed to interact with application

developers who will use the resulting libraries. A working knowledge of

C or C++ programming and experience with software-library development

would be useful.

The Parallel Computing Sciences Department maintains research programs

in a variety of areas, including robust automatic mesh generation,

parallel algorithms, adaptive finite element methods, domain partitioning

and load balancing. Strong collaborations exist between the

Department and other departments at the Laboratories, supporting

research in mathematics and algorithms, computational physics and

engineering, and advanced systems software and tools. A unique

parallel computing environment is available, including a 4500-node

Intel TFlops computer, an 1800-processor Intel Paragon,

a 192-processor SGI Origin system, an 84-processor DEC-8400 system,

and experimental heterogeneous computer platforms.

The post-doc appointment is for a period of one year and may be

renewed for a second year. It includes a competitive salary,

reimbursement of moving expenses, and a professional travel allowance.

Applicants should send a resume and three letters of recommendation to:

Dr. Karen Devine

Parallel Computing Sciences Department, MS 1111

Sandia National Laboratories,

P.O. Box 5800

Albuquerque, NM 87185-1111

The submission date for applications is December 5, 1997,

however applications will be considered until the position is

filled. The position will commence early in 1998.

For further information, contact

Dr. Karen Devine

(505) 845-7585

kddevin@cs.sandia.gov.

Equal Opportunity Employer M/F/V/H. U.S. Citizenship is Required.

------------------------------

From: K. Chen <K.Chen@liverpool.ac.uk>

Date: Thu, 13 Nov 1997 16:03:25 GMT

**Subject: Research Position at University of Liverpool**

UK Engineering and Physical Sciences Research Council

EPSRC Ph.D Studentship

(suitable for mathematics/computing/physics graduates)

A brief description is given below for a planned 3-year project. However

depending the interests of suitably qualified candidates, it is possible

to work on a different project in the general area of applied

mathematical modelling and techniques.

Title: "Fast iterative methods for adaptive solution of fluid

equations with applications to thermal contact problems"

. This project is concerned with understanding the dynamics of compressible

flows and developing effective numerical methods for solving the

discretized nonlinear systems. It involves close collaboration with

the Shell Research Company in Chester, and some financial support.

. The project, to be supervised by Dr Chen (Liverpool) and Dr Scales

(Shell), provides interesting and varied training in analytical and

numerical methods using state-of-the-art computational facilities. There

is a considerable demand from industry for Ph.D graduates with this

type of research training.

Usual EPSRC eligibility rules apply i.e. UK students - full support,

EU students - only full fees and Non-EU students - no support.

To apply, please send a full CV (including 2 referees) to

Dr K. Chen Tel: 0151-794-4741

Department of Mathematical Sciences Fax: 0151-794-4754

University of Liverpool Email: k.chen@liverpool.ac.uk

Liverpool L69 3BX Home page: http://www.liv.ac.uk/maths/applied

------------------------------

From: M. P. Bekakos <M.Bekakos@aueb.gr>

Date: Wed, 12 Nov 1997 16:05:51 +0200 (EET)

**Subject: Research Positions at Athens University**

Ph.D. Positions at Athens University of Economics & Business

Dept. of Informatics

Advanced Computational Mathematics & Parallel Processing Unit (ACM&PP)

Parallel Algorithms and architectures Research Group (PAaRG)

Athens - Greece

This is the announcement for two Ph.D. candidates positions in the area of :

"Parallel Processing Technology, Neural Network Architectures and

Applications" and

"Parallel Processing Technology and Neural Networks: Study and Performance

Evaluation of Parallel Algorithms for Shared / Distributed Memory

Architectures"

We welcome young candidates (below 30 years of age) with an M.Sc. degree in

a relevant topic and a sound background in Mathematics and experienced in

scientific programming. Research activity and/or professional experience

will additionally evaluated in favour of the candidate.

For more information please contact:

Michael P. Bekakos

PAaRG

Dept. of Informatics

Athens University of Economics & Business

Athens 104 34

Greece

Fax : +301-82 25 790

E-mail : mpb@aueb.gr

------------------------------

From: Richard Sincovec <sincovec@ringer.cs.utsa.edu>

Date: Sun, 9 Nov 1997 13:23:45 -0600 (CST)

**Subject: Student Positions at University of Texas at San Antonio**

University of Texas at San Antonio

Ph.D. Computer Science Fellowships

and

M.S. and Ph.D. Research Assistantships

In an effort to recruit outstanding computer science graduate

students, the University of Texas at San Antonio (UTSA) has allocated

funds for a number of fellowships. These fellowships provide full

tuition and fees plus a stipend of $15,000. There are also a number

of graduate student research assistant positions supported by

federally funded research projects in computer systems. Particular

interests are in parallel and distributed computing systems and

applications, parallelizing compilers, high-speed networks,

performance evaluation and simulation.

The Division of Computer Science has 13 faculty and about 480

undergraduate and 60 graduate students (MS and PhD) and is expected to

grow significantly in the next few years. In addition to excellent

general purpose computing and networking facilities, the Division

supports clusters of high-performance and SMP workstations on

high-speed networks for research use. Additional information about the

academic programs and the faculty and their research can be found at

the URL http://www.cs.utsa.edu.

Applicant must have at least a BS degree in computer science or

related disciplines and should be eligible for admission as an M.S. or

Ph.D. degree seeking student in the Division of Computer Science, The

University of Texas at San Antonio.

Prospective students are requested to send a letter of interest, with

a copy of their resume to :

Kleanthis Psarris

Division of Computer Science

The University of Texas at San Antonio

San Antonio, TX 78249-0667

Email: psarris@ringer.cs.utsa.edu

Email communication is preferred. Minority and women students are

especially encouraged to apply.

Richard Sincovec, Director Telephone: 210-458-4434

Division of Computer Science Fax: 210-458-4437

The University of Texas at San Antonio e-mail: sincovec@cs.utsa.edu

6900 North Loop 1604 West

San Antonio, Texas 78249-0677

------------------------------

From: Hans Schneider <hans@math.wisc.edu>

Date: Tue, 11 Nov 1997 13:25:50 -0600 (CST)

**Subject: Contents, Linear Algebra and Its Applications**

Journal Code : 7738

Journal : Linear Algebra and Its Applications

Volume issue : 270/01-3

Year : 1997

Anticipated Publication Date: 26-NOV-97

pp. 1-13

Bounds On Eigenvalues And Chromatic Numbers

D Cao

pp. 15-27

Spectra of Multilevel Toeplitz Matrices: Advanced Theory via Simple Matrix

Relationships

EE Tyrtyshnikov

pp. 29-44

Eigenvalue inequalities and equalities

Horn

pp. 45-77

Generalized Cauchy-Vandemonde matrices

Heinig

pp. 79-108

A sorted partial Jacobi Method and its convergence analysis

H Zha

pp. 109-129

On the extreme eigenvalues of Hermitian (block) Toeplitz matrices

S Serra

pp. 131-154

Real perturbation values and real quadratic forms in a complex vector space

B Bernhardsson

pp. 155-169

Actions that characterize

B Chalmers

pp. 171-189

A refined iterative algorithm based on the block Arnoldi process for large

unsymmetric eigenproblems

ZHONGXIAO Jia

pp. 191-214

Mixed Dominating Matrices

WALTER Morris

pp. 215-229

Matrix Analysis and the Friedrichs Operator of a Quadrature Domain

MIHAI Putinar

pp. 231-253

Locally Polyhedral Linear Inequality Systems

MIGUELA Goberna

pp. 255-273

Finite Metric Spaces of Strictly Negative Type

POULG Hjorth

pp.. 275-286

A stronger extention of the hardy inequality

LH Wang

pp. 287-313

Relationships between discrete-time and continuous-time algebraic Riccati

inequalities

YS Hung

pp. 315-321

A relative perturbation bound for positive definite matrices

Mathias, K Veselic

pp. 323-349

Equality of higher numerical ranges of matrices and a conjecture of Kippenhahn

on Hermitian pencils

CK Li

pp. 351-357

Non negative elements of subgroups of Zn

JC Rosales

------------------------------

From: Ake Bjorck <akbjo@mai.liu.se>

Date: Wed, 12 Nov 1997 08:24:58 +0100 (MET)

**Subject: Contents, BIT**

CONTENTS BIT Volume 37, No. 4 (December 1997)

ISSN 0006-3835

A bound on the maximum strong order of stochastic Runge-Kutta

methods for stochastic ordinary differential equations

K. Burrage, P. M. Burrage, and J. A. Belward, pp. 771--780

Multigrid and multilevel methods for quadratic spline collocation

C. C. Christara and B. Smith, pp. 781--803

On summation formulas due to Plana, Lindel\"of and Abel, and related

Gauss-Christoffel rules, II

G. Dahlquist, pp. 804--832

Pseudospectra for matrix pencils and stability of equilibria

J. L. M. van Dorsselaer, pp. 833--846

Quasi-Monte Carlo methods for numerical integration of multivariate

Haar series

K. Entacher, pp. 846--861

Estimation of optimal backward perturbation bounds for the linear

least squares problem

R. Karlson and B. Wald\'en, pp. 862--869

$NP$-stability of Runge-Kutta methods based on classical quadrature

T. Koto, pp. 870--884

Preconditioning of elliptic problems by approximation in the

transform domain

M. K. Ng, pp. 885--900

Low degree rational spline interpolation

P. Oja, pp. 901--909

Backward stability of a pivoting strategy for sign-regular linear

systems

J. M. Pe\~{n}a, pp. 910--924

The implicit application of a rational filter in the RKS method

G. de Samblanx, K. Meerbergen, and A. Bultheel, pp. 925--947

Fair upper bounds for the curvature in univariate convex

interpolation

J. W. Schmidt and W. Hess, pp. 948--960

On the weighting method for least squares problems with linear

equality constraints

G. W. Stewart, pp. 961--967

On singular points of quasilinear differential and differential-

algebraic equations

J. Tuomela, pp. 968--977

Multi-parameter error resolution for the collocation method of

Volterra integral equations

A. Zhou, pp. 978--987

Publisher's announcement, p. 988

Acknowledgements, p. 989

Index, pp. 990--993

------------------------------

From: Corry Magrijn <magrijn.secsup@tip.nl>

Date: Fri, 14 Nov 1997 13:09:58 +-100

**Subject: Contents, Mathematics of Control, Signals, and Systems**

TABLE OF CONTENTS

Mathematics of Control, Signals, and Systems (MCSS)

Volume 10, Number 2

D. Hinrichsen and J. O'Halloran,

Limits of generalized state space systems under

proportional and derivative feedback,

MCSS 10 (1997), 97-124.

S. Weiland and A. Stoorvogel,

Rational representations of behaviors:

Interconnectability and stabilizability,

MCSS 10 (1997), 125-164.

R.K. Prasanth and M.A. Rotea,

Interpolation with multiple norm constraints,

MCSS 10 (1997), 165-187.

S. Battillotti,

Universal controllers for robust control problems,

MCSS 10 (1997), 188-202.

------------------------------

End of NA Digest

**************************

-------