Interval session at IMACS/ACA'2000

                           Special Session on 
                INTERVAL AND COMPUTER-ALGEBRAIC METHODS 
                       IN SCIENCE AND ENGINEERING
                               at the 
                     6th International Conference on 
                    Applications of Computer Algebra 
                              IMACS-ACA'2000, 
                             June 25-28, 2000,
                          Saint Petersburg, Russia

PREAMBLE: 

There is a need to combine methods of computer algebra and of interval
computations. 

First: 
* Most applications of computer algebra and symbolic computations, 
  (in particular, most applications to control, dynamical system analysis,
  computer graphics, etc.), deal with situations in which we know the
  exact all the coefficients of the corresponding analytical expressions. 
* In many real-life situations, however, these
  coefficients have to be determined from measurements and observations.
  Since a measurement is never 100% accurate, after measuring a value x,
  we can only conclude that the actual (unknown) value of the measured
  coefficient lies within the interval [x-D,x+D], where D is the upper
  bound on the measurement error (guaranteed by the manufacturer of the
  measuring instrument). 
We therefore need to take this interval uncertainty into consideration.

Second: 
* Most algorithms developed in computer algebra assume that all
  the coefficients are (exactly represented) real numbers. 
* In the computers, many real numbers can only be approximately
  represented. The resulting rounding errors lead to the inaccuracy of
  the coefficients in the final result. 
It is therefore desirable to estimate this inaccuracy. For this
estimation, we can also use methods of interval computations. 

In recognition of this need, in 1994, an International Conference on
Interval and Computer-Algebraic Methods in Science and Engineering
(Interval'94) was held in St. Petersburg, Russia. This first
conference of this type was a huge success. At this conference,
more than 100 researchers from 18 countries presented their
practical and theoretical results. 

Since 1994, there has been a tremendous progress both in
computer algebra and in interval computations. This progress is
largely due to the rapidly increasing computer processing speed, which
makes previously theoretical algorithms of computer algebra 
practically feasible. In some cases, we can directly apply these
algorithms; in most cases, however, there is a need for further
fine-tuning, a need which leads to interesting challenging 
new theoretical problems whose solution, in its turn, results in new
exciting applications. 

We believe that time is ripe for a new major meeting devoted to the
relation between computer algebra and interval computations. This
meeting will hopefully not only highlight the results, but it will 
also give a new boost to a much-needed combination of numerical and
symbolic techniques. 

SCOPE: 

For this special session, we are soliciting papers in the following
areas:
* combining symbolic methods and interval computations in solving
  mathematical problems such as solving nonlinear systems of
  equations/inequalities etc.;
* applications of combined interval-analytical techniques in science
  and engineering (and in other possible application areas); 
* special languages, software and hardware tools which either 
  * combine interval techniques with techniques of computer algebra,
    or
  * enhance such a combination;
* theoretical foundations for combining interval and symbolic algebra
  techniques, including (but not limited to):
  * the use of analytical transformations (and other techniques from
    computer algebra) in interval computations;
  * algebraic approach to interval mathematics (including
    interval-based formalisms of computer algebra); 
  * computational complexity analysis of symbolic computation
    problems with interval uncertainty; 
* new semi-heuristic ideas on how interval and computer algebra
  methods can be combined (either in general, and with some specific
  application area in mind), and 
* new potential applications area for the combined interval-analytical
  techniques. 

In this solicitation, we are targeting researchers and practitioners
from both communities: interval computations and computer algebra. 
We are also very much interested in researchers from Constraints
Programming community, who use techniques from interval computations
and/or computer algebra. To
achieve a greater success, we are making this appeal as broad as
possible:
* It is OK to have a result which is mainly devoted to interval
  computations, but has some relation to computer algebra.
* It is also OK to have a result which is mainly devoted to 
  computer algebra, but has some relation to interval computations. 

Since this session is oriented towards several different communities, we
encourage the authors to do their best to be understandable by
researchers from all these communities (even if this means adding 
extra phrases into the introduction which, e.g., for an interval
computations community would not be necessary at all). 

SUBMISSION:

A 2-page abstract in LaTeX format must be submitted,
by May 15, 2000 to both organizers (by email, if possible): 

V. M. Nesterov
P. O. Box 52
St. Petersburg 256
195256 Russia
email nest@into.nit.spb.su

Vladik Kreinovich
Department of Computer Science
University of Texas at El Paso
El Paso, TX 79968, USA
email vladik@utep.edu

Abstracts will be published before the conference.

We are also currently planning to publish refereed proceedings as 
a special issue of the "Reliable Computing" journal. 

If you are interested in coming, please let the organizers know ASAP, 
and, if possible, send a title (even preliminary title if necessary) 
of your submission.

VENUE:

St. Petersburg is a majestic old imperial capital of Russia.
Numerous imperial palaces located in the city and in the suburbs
attract millions of visitors every year. The main of these palaces -
Winter Palace - hosts the Hermitage, one of the world
largest art museums. Fortresses, churches, mansions, theaters,
monasteries line up the beautiful streets and embankments of this city
of 100+ islands nicknamed the Northern Venice. 

June is the most beautiful and romantic time to visit St. Petersburg, 
the time of the world-famous White Nights, when the sun hardly hides 
behind the horizon, and you can not find even the brightest stars 
in the night sky. All night long street musicians play traditional 
Russian music and new Russian rock, and love-struck young people roam 
the streets in which the palaces and bridges, barely visible in the
dimmed light (and sometimes also shielded by fog and/or drizzle), 
create the magic atmosphere of unreal fairy tale. 

Although the city is located at the same latitude as Alaska, Gulf
Stream makes it much warmer; in June, an average temperature is about
20 C (70 F). 

For information about hotels, registration fees, visas, etc., see 
the IMACS-ACA'2000 website.

[<--] Back to Forthcoming Conferences on Interval Computations

[<--] Back to the main menu of the Interval Computations website