NA Digest Monday, December 18, 2006 Volume 06 : Issue 51

Today's Editor:
Tamara G. Kolda
Sandia National Labs

Submissions for NA Digest:

Mail to

Information via email about NA-NET:

Mail to


From: Jens Hugger <>
Date: Tue, 12 Dec 2006 08:08:16 -0500
Subject: Re: Fortran versus Matlab in teaching numerical analysis

Professor Martin Gander wrote in NA Digest V. 06, # 50 about Fortran versus
Matlab in teaching numerical analysis.

I am using both. Actually it is Maple and C, but essentially the same thing.
Maple, Mathematica, Matlab and so on all offer "easy access".
In most Num.An. courses you suffer from "lack of prerequisites". In my
ODE-course for example, we do not have the time to cover solution of linear
equation systems. Hence we treat this as a black box and use Maple's built in
linear equation solver.
Also in a very short time, the students are able to graphically relay their
results with Maple.
Obviously, you very soon run into problems with excessive computing times, but
Maple is generally enough for the "First course in..."
For the more advanced courses I then introduce external C-routines called from
inside Maple.
I still believe that Fortran is the best language for Numerical Analysis, but
unless you can count on a significant number of students taking more than just
one or two courses, I think that it is wise to built on the prior knowledge
among the students. At University of Copenhagen we use Maple for the
introductory calculus courses, and C is still the most dominant "real language".

Jens hugger
Assoc. Prof. in Numerical Analysis
University of Copenhagen


From: "Komzsik, Louis" <>
Date: Mon, 11 Dec 2006 08:33:51 -0800
Subject: MATLAB versus FORTRAN in teaching Numerical Analysis

It is my personal belief that there is significantly more educational
value in requiring a student to implement a certain numerical algorithm
in FORTRAN than just being able to use some canned numerical algebraic
components of MATLAB, latter being an unfortunately dominant trend in
undergraduate numerical methods courses in many state universities.
Considering also the fact that there is a serious shortcoming of new
computer scientists and engineers who know FORTRAN and, as Prof Gander
pointed out, there are widely spread software tools (like NASTRAN) with
multi-million lines of operational FORTRAN code, teaching numerical
methods classes with FORTRAN could help in this regard as well.

Louis Komzsik
Chief Numerical Analyst
UGS Corp.


From: Joe Grcar <>
Date: Tue, 12 Dec 2006 10:38:58 -0800
Subject: Programming in Teaching Numerical Analysis

Regarding Martin's question about what programming
language to use in teaching numerical analysis: the
best ones are whichever instructors know well and
whichever best serve to illustrate simple concepts,
hence probably matlab.

However, mastery of interpreted language environments
of the matlab kind emphatically does not prepare students
to function in the real world of scientific computing.
There it is necessary to know both the c++ and f90
families as well as scripting languages such as perl and

An aspect of computational science that I find lies
outside the experience of academic numerical analysts is
that most codes are too complex and too large for any
one individual to understood completely. It is not enough
to master conceptually simple mathematical algorithms,
one has to master how those algorithm are pieced together
through control and data structures that cannot be changed
(or can only be changed through negotiation with many
other parties) and which are made all the more complex
by parallelism. A particularly good recent paper that
illustrates the overlap of applied math, computer science,
and domain science in one application code is

This description of new software for protein folding was
selected as the best paper at this year's supercomputing
conference. My question to Martin and other teachers is:
how does one prepare students to function as a numerical
analyst in the teams of a dozen or so people who write
this kind of software?

Best holiday wishes from a rainy San Francisco, -- Joe

Joseph Grcar
Lawrence Berkeley National Laboratory


From: Mike Sussman <>
Date: Wed, 13 Dec 2006 18:35:22 -0500
Subject: Re: Matlab versus Fortran in Teaching Numerical Analysis

In response to Martin Gander's question, my opinion is that Matlab is
by far the best language to teach numerical analysis.

I have taught a graduate level numerical analysis laboratory in the
Math department at the University of Pittsburgh for several years.
The focus of the lab is primarily numerical analysis and secondarily
on programming. It seems to me that if you want to teach numerical
analysis, then Matlab greatly simplifies the associated programming
task. In the following four paragraphs, I list some of the items I
feel that strongly point toward Matlab over Fortran.

Writing a Fortran (or C or C++) program involves variable declarations
that get in the way of numerical work. Why, for example, should the
students have to be taught the difference between an integer, a
single-precision real number, a double-precision real number, a
single-precision complex number, or even (if you are lucky) a
double-precision complex number? And then there is the difference
between a scalar and array variable. Wouldn't the time spent on
these distinctions be better spent discussing, to name one, roundoff

One of the most powerful teaching tools is graphical output. Matlab
makes it easy to present output graphically, but you will have to use
some plotting package to generate graphical output from more
general-purpose languages. You will need to explain how to call the
plotting routines, and how to link against the required library. I
imagine you will spend about a class hour explaining this material--is
it worth the time?

Students will have to learn to debug their code. Debugging is the
hardest and most time-consuming of the code generation tasks. Matlab
provides an integrated window-based debugger, while you will need
to locate and install one for compiled languages such as Fortran.

It is often the case that students will have five or more subroutines
or functions grouped into a single logical program. A root-finder
along with several examples comes to mind. To do this "right" in a
command-line environment involves a "make" or similar program. In an
integrated development environment, the association among the several
files must be described to the environment. Explaining how to do this
takes valuable class time.

On the other hand, are you short-changing the students by not exposing
them to the workhorse programming languages? My feeling is that
students who progress from Matlab to a mainstream programming language
do have some trouble dealing with the details that Matlab hides (for
example, declarations). But many of the students will never progress
to another programming language. In today's environment of very
powerful computers, a vast amount of numerical work can be done on
computers running Matlab. There is a tradeoff to be made, and it is
based on the number of students who are going to be learning Fortran

In summary, for a class focussed on numerical analysis Matlab is
a better choice of programming language than mainstream programming
languages such as Fortran, C, Java, etc.

Mike Sussman


From: "Dr. Pijush K Bhattacharyya" <>
Date: Mon, 11 Dec 2006 18:13:12 +0000
Subject: Teaching of Numerical

Martin Gander wrote:

"Following Gene Golub's suggestion for discussions on the NA-Digest, I would
like to ask how people feel about programming languages to
teach numerical analysis. Should one use Fortran, one of the top 10 algorithms
of the 20th century, a language in which most large scale existing codes are
written ? Or should one use Matlab, or even C++ or Java ? "

I see no reason to believe that the teaching of Num Analysis is greatly affected
by the choice of its usage of computer language. In fact quite often I have used
Matlab, C++ and Fortran , depending on the student body I'm facing. In the
eighties, as a visiting faculty at UC-Berkeley, I taught Num Analysis both to
2nd year math students in C++ and also to ph.d qualifiers of Mech Eng in
Fortran - without worrying much, as the cohorts were used to these two
languages respectively.

Of late, I'm using Matlab to teach the subject to a mixed bag of engineers!
So I'm not totally convinced of the choice in the usage of programming
languages. But then, I could be wrong!

Dr. Pijush K Bhattacharyya
Senior Professor & Director of International Operations
School of Informatics
University of Westminster


From: Tim Davis <>
Date: Mon, 11 Dec 2006 11:05:54 -0500
Subject: Re: MATLAB vs Fortran

My bias is MATLAB for teaching numerical analysis, and C/C++ where needed. In
a recent book on sparse direct methods, for example, I chose C to get into the
detailed algorithmic/data structures issues inside sparse direct methods, but
MATLAB to clearly illustrate overall algorithms and methods. C is clear,
fast, and expressive. MATLAB is superior to C, except when you need to
express computations that don't fit well into a pure matrix formulation
(computing an elimination tree, traversing a graph, ...).

Regarding Martin's comment that Fortran is "a language in which most large
scale existing codes are written": for one domain (sparse direct methods), I
included a nearly exhaustive list of available software in my book, sequential
and parallel, commercial and otherwise. Some of these are very large-scale
codes. (see Skimming that
table, I see 21 codes in Fortran, and 24 in C/C++. I could name many other
large scale codes written in C/C++ as well. For example, I would hazard a
guess that most optimized BLAS are written in C (Intel MKL, AMD's ACML, etc).
Excluding m-files that are part of MATLAB, the core of MATLAB itself is mostly
written in C/C++, with some Java (GUI's and such), and some Fortran (LAPACK).

(sorry - this isn't meant to be a plug for my book. It's just one data point
I'm familiar with).


From: Chandler Davis <>
Date: Sun, 17 Dec 2006 13:28:29 -0500
Subject: Olga Taussky Todd Lecture Award to Pauline van den Driessche

Olga Taussky Todd Lecture Description:

AWM and EWM, together with the organizers of the ICIAM 2007 Congress,
will award th Olga Taussky Todd Lecture at the ICIAM 2007 Congress in

This honor is to be conferred to a woman who has made outstanding
contributions in applied mathematics and /or scientific computation. The
name of this lecture pays tribute to the memory of Olga Taussky Todd,
whose scientific legacy is in both theoretical and applied mathematics,
and whose work exemplifies the qualities to be recognized.

The Committee has chosen Pauline van den Driessche as the Lecturer. She
has accepted and will deliver the Lecture at ICIAM 2007. A fuller press
release will be issued by AWM and EWM.

Chandler Davis


From: Marja-Leena Rantalainen <>
Date: Fri, 15 Dec 2006 14:14:11 +0200 (EET)
Subject: Call for submissions, EUROGEN 2007, Finland, June 11-13 2007

EUROGEN 2007 conference
On Evolutionary and Deterministic Methods for Design,
Optimization and Control with Applications to Industry and Social Problems

will take place in Jyvaskyla, Finland, June 11-13, 2007.

Algorithmic topics
1. Evolutionary algorithms (including Evolutionary programming, Evolution
Strategies and Genetic Algorithms)
2. Multi-objective evolutionary algorithms and constraint handling
3. Hybridization with standard optimization tools (gradient based methods,
combinatorial optimization methods)
4. Response Surface methods and other embeddings of Machine learning
5. Parallel and distributed Evolutionary Algorithms (from LANs to GRID)
6. Multi-disciplinary optimization methods
7. Economic and Social simulations
8. Multi criteria decision making tools

Deadline for abstract submission is January 15, 2007.

For more information about the conference, please visit the Website

More information about Jyvaskyla area could be found at


From: Pat Keast <>
Date: Tue, 12 Dec 2006 12:46:04 -0500
Subject: Summer School, Jul 15 - Aug 10, 2007, Dalhousie, Halifax, Canada

The sixth annual Summer School of the Atlantic Association for Research in the
Mathematical Sciences (AARMS) will be held in the Department of Mathematics
and Statistics on the campus of Dalhousie University, Halifax, Nova Scotia,
from July 15 to August 10, 2007. The School Director is Pat Keast, email .

The summer school is intended for graduate students and promising
undergraduate students from all parts of the world. Each participant is
expected to register for two of the courses. Each course consists of five
ninety-minute lectures each week. These are Dalhousie University graduate
courses and we will facilitate transfer credit to the extent possible.

As in the previous Schools, four courses will be offered:
* Polynomials
Instructor: Ed Barbeau, University of Toronto
* Statistical Numerical Integration
Instructor: Alan Genz, Washington State University
* Mathematical Models in Ecology and Evolution
Instructor: Frithjof Lutscher, University of Ottawa
* Introduction to Number Theory
Instructor: Alf van der Poorten, MacQuarie University

If your application is accepted, your accommodation and meals will be covered
while the School is running. Unfortunately, we cannot cover travel costs. For
more information on course content see To
obtain an application form contact or download a form
from the School website.


From: gerhardwilhelm weber <>
Date: Wed, 13 Dec 2006 09:24:24 +0200
Subject: EURO XXII 2007, Prague, July 8-11, 2007 (and satellite events)

Prague, Czech Republic, July 8-11, 2007

The 22nd European Conference on Operational Research
EURO XXII is organized by the Czech Society for Operational Research
in co-operation with the University of Economics,
Prague and will be held on the University campus site located
near to the Historic centre of Prague.

On-line registration: starts in October 2006
Deadline for abstract submission: February 28, 2007
Deadline for early registration: April 15, 2007
Conference: July 8-11, 2007

We invite all researchers, academicians, practitioners, as well
as students interested in all branches of operational research,
mathematical modeling and economic analyses to participate
at the conference and to present their papers.

Invited and contributed papers will be organised in parallel
sessions of 90 minutes, with three or four papers in each session.
Abstracts can be submitted via the Conference web page - Every participant can present no more than
one paper at the Conference.

Invited and contributed sessions will generally be a part of
Conference main streams. Their list together with two
responsible organisers is presented on the Conference web
page. The researchers who want to organise an invited session
or contribute a paper within an invited session should contact the
stream organiser or PC chair with their proposals.

There are some scientific events in preparation already,
organized by EURO Working Groups and their friends:

A Joint EUROPT-OMS Meeting:
2nd Conference on Optimization Methods & Software and
6th EUROPT Workshop "Advances in Continuous Optimization",
Prague, Czech Republic, July 4-7, 2007,

EURO-CBBM Workshop -
Workshop on OR in Computational Biology, Bioinformatics and Medicine,
Prague, Czech Republic, July 7-8, 2007,

A Workshop on OR for Developing Countries ---
Young Researchers and PhD Symposium


From: Meshfree Methods <>
Date: Mon, 18 Dec 2006 13:17:03 +0100
Subject: Meshfree Methods for Partial Differential Eqns, Germany, Sep 17-20, 2007


Fourth International Workshop
Meshfree Methods for Partial Differential Equations

Bonn, Germany

September 17 - 20, 2007


Prof. Ivo Babuska (University of Texas, Austin, USA)
Prof. Ted Belytschko (Northwestern University, USA)
Prof. Michael Griebel (Universitaet Bonn, Germany)
Prof. Antonio Huerta (Universitat Politecnica de Catalunya, Spain)
Prof. Wing Kam Liu (Northwestern University, USA)
Prof. Harry Yserentant (Technische Universitaet Berlin, Germany)

Important Dates and Deadlines:

May 1, 2007 Early Registration and Abstract submission
August 1, 2007 Confirmation and Program



Date: Mon, 11 Dec 2006 10:31:58 -0500
Subject: 2nd CFP: Athens-HERCMA 2007 Conference, Sept. 20-22, 2007


20-22 September 2007, Athens, Greece.

Framework-Objectives: The 8th Hellenic European Research on Computer
Mathematics & its Applications (HERCMA) Conference will be held in the Athens
University of Economics & Business (AUEB) on September 20-22, 2007, in
Athens, Greece. The main theme within the Conference will be Computer
Mathematics and its Applications and special emphasis will be given to
Computational Mathematics, High Performance Computing, Operational Research
and Statistics, Mathematics in Economics and Industry.

Call for Papers: Papers on all aspects of Computer Mathematics and Scientific
Computing are solicited.

Schedule: Updated information about HERCMA 2007 may be retrieved from our web
site at

E.G. COFFMAN, Columbia University-NY, U.S.A.
H. FOLLMER, Humboldt University, Germany
Sir C.A.R. HOARE, FRS, Oxford University, England
J. KEVREKIDIS, PACM-Princeton University, U.S.A.
P.-J. LIONS, College de France, France
T. SATO, The Earth Simulator Center, Japan
K. SUGIHARA, University of Tokyo, Japan
A. TOLSTYKH, Russian Academy of Sciences &(MIPT), Russia

Important Dates:
Deadline for the submission of the full papers/extended abstracts:
20 February, 2007
Deadline for mini-symposia proposals: 20 February, 2007
Notification of acceptance: 10 April, 2007
Deadline for the submission of the complete papers: 30 May, 2007
Deadline for the early payment: 30 May, 2007

For further information please look at the HERCMA home page at Web site:


From: Alex Pothen <>
Date: Sat, 16 Dec 2006 05:33:02 -0500
Subject: CSC07 Workshop Dinner in honor of Iain Duff on Feb 17, 2007

The SIAM Workshop on Combinatorial Scientific Computing (CSC07), being held in
Costa Mesa, CA from February 17 to February 19, 2007, preceding the SIAM
Conference on Computational Science and Engineering, is now open for
registration. The titles and abstracts of invited and accepted talks are
available from the workshop web page,

Poster presentations will be accepted until January 22, 2007.

A workshop dinner is being organized for Saturday, Feb 17, 2007 at the Hilton
Orange County/Costa Mesa, when we will help one of our invited speakers, Iain
Duff, celebrate his sixtieth birthday! Details about the dinner are available
at the workshop web page, and those interested in participating should contact
the organizers at, before January 22, 2007.

We have received support from the Department of Energy for partial travel
support of students and early career researchers, and some funds are still
available for such support. Information on applying for support is available
from the workshop web page.

Alex Pothen
for the CSC07 Program Committee


From: Erik Burman <>
Date: Mon, 18 Dec 2006 09:54:09 -0500
Subject: VMS2007, Miniworkshop, Stabilized Methods, EPFL, Lausanne, Feb 12-13

A miniworkshop on variational multiscale methods and stabilized finite element
methods will be held at the Institute of Analysis and Scientific Computing
(IACS), EPFL, Lausanne on the 12th and 13th of February.

Confirmed participants include:

M. Behr, RWTH Aachen
M. Braack, Christian-Albrechts-Universität zu Kiel
R. Codina, Universitat Politècnica de Catalunya, Barcelona
A. Ern, ENPC Paris
J.-L. Guermond, Texas A&M University, College Station
P. Hansbo, Chalmers Tekniska Högskola, Göteborg
T. Hughes, University of Texas, Austin
V. John, Universität des Saarlandes, Saarbrücken
M. Larson, Umeå universitet
G. Lube, Universität Göttingen
G. Sangalli, Università degli Studi di Pavia
L. Tobiska, Otto-von-Guericke Universität Magdeburg
S. Turek, Universität Dortmund
J. van der Vegt, Universiteit Twente, Enschede

Those interested are encouraged to contact
for information or registration. A limited number of places are available.

For further information see also the web-site:

On behalf of the organizing committee,

Erik Burman


From: "M.Todorov" <>
Date: Mon, 11 Dec 2006 11:51:33 +0200
Subject: Applications of Math in Eng & Econ, AMEE'07, Bulgaria, Jun 8-14, 2007

Announcing conference AMEE'07, Jun 08-14, 2007

33rd International Conference "Applications of Mathematics in
Engineering and Economics" (AMEE'07) organized by the Faculty of
Applied Mathematics and Informatics, Technical University of Sofia,

The main goal of the conference is to bring together experts and young
talented scientists from Bulgaria and abroad, to discuss the modern trends,
and to ensure exchange of views in various applications of mathematics in
engineering, physics, economics, biology, etc. It will be subject to the motto
"Nonlinear phenomena - mathematical theory and environmental reality". The
Organizing Committee encourages the taking part of students and postgraduates
in the sessions of the Conference. The keynote speakers are supposed to
organize special sessions. The Organizing Committee stipulates a special
discussion concerning the development and use of software innovations in the
scientific computing and the student training in this area.

TOPICS: Potential Theory and Partial Differential Equations; Mathematical
Analysis and Applications; Differential Equations and Differential Geometry;
Numerical Methods in Mathematical Modeling; Algebraic Methods in Informatics;
Software Innovations in Scientific Computing

A.Haghighi (USA), H.Kojouharov (USA), M.Konstantinov (Bulgaria), R.Lazarov
(USA), A.Loskutov (Russia), P.Minev (Canada), A.Schmitt (Germany), P.Velmisov

For more detailed information please visit the conference URL: or contact:


From: Iain Duff <>
Date: Fri, 15 Dec 2006 18:58:04 GMT
Subject: CERFACS Anniversary Meeting, Oct 11-12, 2007

Diary date for CERFACS friends

CERFACS will be celebrating its 20th Anniversary on October 11-12. There
will be general CERFACS activities on Thursday 11th and the Parallel
Algorithms Team will have a one-day meeting on Friday 12th. More details
will be available early next year but you should note the dates in your
diary now.

Iain Duff ( and Serge Gratton (


From: Padma Raghavan <>
Date: Mon, 11 Dec 2006 11:24:15 -0500
Subject: Faculty Position in Scientific Computing at Penn State

The Pennsylvania State University
Department of Computer Science and Engineering
Faculty Position
Complete Announcement at:

Applications are invited for several tenure-track faculty positions
at all ranks. Outstanding candidates in all areas of computer science
and engineering will be considered. One of the areas of particular
interest is: Scientific Computing with a focus on scalable algorithms
and systems for modeling and simulation. Candidates are also sought
for interdisciplinary joint positions in the areas of bio-inspired
computing and nano-technologies, computational science and engineering,
and bioinformatics.

The Department of Computer Science and Engineering (CSE) has
31 collegial, tenure-track faculty. The undergraduate and Doctoral
programs are ranked in the top 20 by the US News and World Report.
The university is committed to growing the faculty ranks over the
next several years.

Penn State is a major research university and is ranked third in
the nation in industry-sponsored research among universities. US
News and World Report consistently ranks PSU's College of Engineering
undergraduate and graduate programs in the top 15 of the nation.
The university is located in the beautiful college town of State
College in the center of Pennsylvania.

Applicants should hold a Ph.D. in computer science, computer
engineering, or a closely related field and should be committed to
excellence in both research and teaching. Applications should be
received by January 31, 2007 to receive full consideration.

To apply by electronic mail, send your resume (including curriculum
vitae and the names and addresses of at least three references) as
a postscript file or pdf file to

For more information about the Department of CSE at Penn State, see Click the Faculty link under Employment.

Penn State is committed to affirmative action, equal opportunity
and the diversity of its workforce. Women and minorities are
encouraged to apply.


From: Roy Mathias <>
Date: Fri, 15 Dec 2006 17:29:49 -0500
Subject: Lectureship/Senior Lectureship in Comp. Math., U. Birmingham (UK)

The School of Mathematics at the University of Birmingham will
be appointing a

Lecturer/Senior Lecturer in Computational Mathematics
(salary 32,471-49,116 GBP, approx. $63,400-$95,849)

NOTE Application Deadline: Jan 15 2007

This is part of an on going strategy developing an area of strength in
computational mathematics. Applicants in Numerical Linear Algebra
and Optimisation are particularly sought, but all areas of computational
mathematics and statistics will be considered.

The School of Mathematics was ranked 5th nationally among Mathematics
Departments in the latest Times Higher survey.

Faculty in numerical analysis and optimisation include Roy Mathias,
Michal Kocvara (Jan 2007), Daniel Loghin, Natalia Petrovskya,
Joerg Fliege, Peter Butkovic, Sandor Nemeth. At least one additional
appointment in optimisation will be made.

Inquiries to Roy Mathias (

Further information about the School of Mathematics can be found at

Further information about the position can be found at


From: Monika Nitsche <>
Date: Thu, 14 Dec 2006 14:30:21 -0500
Subject: Position in mathematical biology/biostats, University of New Mexico

Assistant Professor position in Biology or Math/Statistics in a probationary
appointment leading to tenure decision. Joint appointment. PhD in Biology,
Mathematics, Statistics or related discipline required by starting date.
Research interests in mathematical biology, computational biology or
biostatistics. For best consideration, submit complete application materials
by January 16, 2007 to Yvonne Martinez-Ingram, Math/Statistics Search
Committee, UNM Biology Department, MSC03 2020, 1 University New Mexico,
Albuquerque, NM 87131. For complete job and application requirements see:

The University of New Mexico is an Equal Opportunity/Affirmative Action
employer and educator.


Date: Tue, 12 Dec 2006 12:06:35 -0800
Subject: 3 positions in Appl/Comput Math at UNLV

The Department of Mathematical Sciences at the University of Nevada, Las Vegas,
invites applications for a full-time, 9-month, tenure-track Assistant Professor
in Applied/Computational Mathematics, commencing Fall 2007. Three positions are
available (contingent upon funding).

Details can be found at AMS website
and UNLV website

Jichun Li, Associate Professor
Dept. of Mathematical Sciences
Univ. of Nevada, Las Vegas


From: Douglas Heggie <>
Date: Tue, 12 Dec 2006 07:01:23 -0500
Subject: Lectureship in Applied Mathematics at the University of Edinburgh

The School of Mathematics invites applications from highly qualified
candidates for a lectureship in Applied Mathematics. Candidates with
a background in computational mathematics and/or experience with
topics in biology/medicine or materials science are particularly
encouraged to apply.

Salary scale £32,471 - £38,772 (Lecturer, Grade 8).

The position is available from 1 August 2007 or at another date by
mutual agreement.

You will hold a Ph.D. in mathematics or a related discipline, and have
an excellent research record. You will have well formulated research
plans enabling you to contribute strongly to a highly-rated
research-led School. You will also be committed to excellence in
teaching at the undergraduate and postgraduate levels.

The School of Mathematics is committed to vigorous expansion in
computational and interdisciplinary applied mathematics, to build on
the recent appointments of Professor Ben Leimkuhler and Dr Assyr
Abdulle, supported by existing staff.

For further information, see, entering the
vacancy reference 3006760.

The closing date is 2 February 2007.


From: Birsen Yazici <>
Date: Mon, 11 Dec 2006 15:02:13 -0500
Subject: Postdoctoral position at RPI

Post-doctoral position at RPI

As part of a DoD funded project in breast cancer diagnosis and diffuse
optical tomography. Rensselaer Polytechnic Institute is seeking applications
for a post-doctoral research associate position at the Electrical,
Computer and Systems Engineering Department and Mathematics Department.
Qualifications: Ph.D. Degree in in mathematics, theoretical
physics,electrical and computer engineering, computer science or
related disciplines. Expertise in numerical solutions of PDEs, knowledge in
inverse problems, interest in diffuse optical imaging and medical
applications, good computing/programming and communication skills. Position
is for 1 years (potentially renewable). Start date immediately.Interested
applicants please send your resume and references to Dr. Birsen Yazici at RPI has a well-recognized leadership role in the area
of inverse problems and it offers exceptional work environment and
competitive salaries.


From: Mituhiro Fukuda <>
Date: Mon, 18 Dec 2006 04:46:05 -0500
Subject: Post-doctoral opportunity in continuous optimization at Tokyo Tech

Post-doctoral Opportunity in Continuous Optimization
at Tokyo Institute of Technology (Japan)

We are looking for a promising and highly motivated candidate for a
year post-doctoral position in continuous optimization at Tokyo
Institute of Technology, one of the leading university of science and
technology in Japan. This program is financed by the grant "Front-Runner
Nurturing Program" of Global Edge Institute

The candidates must hold or expected to hold a PhD degree in
Optimization, Applied Mathematics, Numerical Analysis, Computer Science,
or Computational Chemistry by the time of employment.
Also, he/she must have a background in continuous optimization or
computational chemistry. A research experience focused on theoretical or
computational optimization is welcome. However, there will be given
priority for candidates who have some experience or interest in
optimization problems related to Semidefinite Programming as well as its
applications. If possible, the candidate is expected to help the
development or the applications of the software SDPA/SDPARA maintained
by our group. It will be excellent if the candidate has further
knowledge of computational quantum chemistry and/or parallel computing.

Starting Date: April 1st, 2007 or later (negotiable)
Duration: Maximum of ONE year until March 31st, 2008 (no extension)
Salary: Negotiable according to the expected potential of the candidate
Research Environment:
The candidate will have opportunities to cooperate with Mituhiro
Fukuda (Global Edge Institute, Tokyo Institute of Technology), Prof.
Masakazu Kojima (Department of Mathematical and Computing Sciences,
Tokyo Institute of Technology), Maho Nakata (Advanced Center for
Computing and Communication, RIKEN) and others experts in optimization
during this period.

The research will be conducted all in English and no Japanese is
required. The Tokyo Institute of Technology is one of the university
which has a high percentage of graduate students and also foreign
students in Japan. It has a professional atmosphere to promote research
within the campus. Also located in a suburb area within minutes of the
cosmopolitan Tokyo.

Deadline for the application:
January 31st, 2007 or until the position is filled
More details for submission at:


From: Raimondas Ciegis <>
Date: Wed, 13 Dec 2006 13:42:33 +0200 (EET)
Subject: Contents, Mathematical Modelling and Analysis vol.11 no.4


The Baltic Journal on Mathematical Applications, Numerical Analysis and
Differential Equations
ISSN 1392-6292, ISSN 1648-3510 online, Electronical edition:

Raimondas {\v C}iegis (Editor) Volume 11, Number 4, 2006 (p. 357-474)


G. Alobaidi, M.C. Haslam, R. Mallier.
Vortices on a Sphere. p.357-364

A.R. Ansari, B. Hossain, B. Koren, G.I. Shishkin.
Robust Numerical Methods for Boundary-Layer Equations for a Model Problem
of Flow over a Symmetric Curved Surface. p.365-378

A. Esi.
On Some New Paranormed Sequence Spaces of Fuzzy Numbers Defined by Orlicz
Functions and Statistical Convergence. p.379-388

A.A. Grigor'ev, V.I. Gromak.
The Power Expansions of the Solutions of the First Painlev$\acute{e}$
Hierarchy. p.389-398

Z. Navickas, L. Bikul\v{c}ien\.{e}.
Expressions of Solutions of Ordinary Differential Equations by Standard
Functions. p.399-412

S. Ogorodnikova, F. Sadyrbaev.
Multiple Solutions of Nonlinear Boundary Value Problems with Oscillatory
Solutions. p.413-426

E. Pais, J. Janno.
Inverse Problem to Determine Degenerate Memory Kernels in Heat Flux with
Third Kind Boundary Conditions. p.427-450

V. Starikovi\v{c}ius, R. \v{C}iegis, A. Jaku\v{s}ev.
Analysis of FV Approximations for Solving Problems of Multiphase Flows in
Porous Media. p.451-474


From: Liz Martin <>
Date: Mon, 11 Dec 2006 10:52:58 +0000
Subject: Contents, Nonlinearity, volume 20, issue 1, January 2007

Volume 20, Issue 1, January 2007

All articles are free for 30 days after publication on the web. This issue
is available at: URL:

Hydrodynamics of bacterial colonies
J Lega and T Passot
(Multimedia files accompany this article)


On a zero speed sensitive cellular automaton
X Bressaud and P Tisseur

Degenerate dispersive equations arising in the study of magma dynamics
G Simpson, M Spiegelman and M I Weinstein

Wavetrain response of an excitable medium to local stochastic forcing
R E Lee DeVille and Eric Vanden-Eijnden

Shadowing by non-uniformly hyperbolic periodic points and uniform
A Castro, K Oliveira and V Pinheiro

Regularity of weak foliations for thermostats
G P Paternain

Subshift attractors of cellular automata
E Formenti and P K\r{u}rka

Jordan domain and Fatou set concerning diamond-like hierarchical Potts
Q Jianyong and G Junyang

Subadditive pressure for triangular maps
A Manning and K Simon

$L^{q}$ spectra and R\'enyi dimensions of in-homogeneous self-similar
L Olsen and N Snigireva

A diffuse-interface approximation for surface diffusion including adatoms
A R\"atz and A Voigt

Minimal coupled cell networks
M A D Aguiar and A P S Dias


From: Chi-Wang Shu <>
Date: Thu, 14 Dec 2006 12:29:09 -0500
Subject: Contents, Journal of Scientific Computing

Journal of Scientific Computing

Volume 29, Number 3, December 2006

Partitions for Spectral (Finite) Volume Reconstruction in
the Tetrahedron
Qian-Yong Chen, pp.299-319.

An Improvement of a Recent Eulerian Method for Solving PDEs
on General Geometries
John B. Greer, pp.321-352.

A Level Set Framework for Capturing Multi-Valued Solutions
of Nonlinear First-Order Equations
Hailiang Liu, Li-Tien Cheng and Stanley Osher, pp.353-373.

Conservative Finite Difference Formulations, Variable
Coefficients, Energy Estimates and Artificial Dissipation
Jan Nordstrom, pp.375-404.

End of NA Digest