The special session on Computer Algebra Systems and Interval Methods at ACA’2023 to be held July 17-21 at the Warsaw University of Life Sciences, Warsaw, Poland.
Organizers (in alphabetic order):
- Milan Hladik, Charles University, Czech Republic.
- Małgorzata Jankowska, Poznan University of Technology, Poland.
- Vladik Kreinovich, University of Texas at El Paso, USA.
- Bartłomiej Jacek Kubica, Warsaw University of Life Sciences — SGGW, Poland.
- Nathalie Revol, Inria, Ecole Normale Superieure de Lyon, France.
- Iwona Skalna, AGH University of Science and Technology, Poland.
Overview:
Interval methods are a class of algorithms that are accurate and even allow to obtain a guaranteed result. They also provide a useful and appropriate tool to describe the uncertainty of parameters, discretization inaccuracy and numerical errors. Nevertheless, they are usually time consuming and memory demanding.
A particular problem encountered in the interval calculus is the so-called dependency problem. While a similar problem has already been observed for floating-point numbers — various formulae lead to various accuracy — in the case of intervals, its importance is even higher. While floating-point errors could usually be assumed to be `small’, intervals can be arbitrarily wide; consequently differences between mathematically equivalent expressions can be huge. Even computing the value of a univariate polynomial is non-trivial when the coefficients are interval-valued (the Horner’s method is not necessarily optimal either with respect to time or accuracy). It makes symbolic preprocessing of expressions particularly important and useful in interval algorithms.
While for interval solvers of polynomial systems, some effort (mostly based on the Gröbner basis theory) has been done by several researchers, non-polynomial systems are definitely under-studied.
The Session is going to provide a forum for interval researchers to share their experiences, and present possible combination and cooperation of various symbolic/computer algebra and interval algorithms.
Topics of interest include (but are not limited to):
- the use of Gröbner basis theory in interval algorithms,
- symbolic and algorithmic differentiation methods,
- term rewriting, substitution, or other formula transformation to optimize the interval calculi,
- symbolic transformations of interval-valued matrices,
- the use of Poisson series processors, and other symbolic methods in cooperation with the interval analysis,
- applications of computer algebra systems for global optimization/equations solving methods,
- applications of computer algebra systems for ordinary and partial differential equations.
If you are interested in proposing a talk, please send an abstract to Bartłomiej Jacek Kubica; the preferred address is:
bartlomiej.jacek.kubica@gmail.com.
Please use this LaTeX template for your abstract and send both the LaTeX source and a compiled PDF version.
We suggest that abstracts be at least half a page including references.
All other rules of the ACA 2023 Conference apply, including the deadlines, etc.
If you have any questions, please contact Bartłomiej J. Kubica, as well.
Talks
Go to:
ACA’2023 main page
Conferences on Applications of Computer Algebra main page.