Loading…
"Interval rational = algebraic" revisited
In [1], it is shown that if we add "interval computations" operation to the list of arithmetic operations that define rational functions, then the resulting class of "interval-rational" functions practically coincides with the class of all algebraic functions. By "practicall...
Saved in:
Published in: | SIGNUM newsletter 1996-01, Vol.31 (1), p.14-17 |
---|---|
Main Authors: | , |
Format: | Article |
Language: | English |
Citations: | Items that cite this one |
Online Access: | Get full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | In [1], it is shown that if we add "interval computations" operation to the list of arithmetic operations that define rational functions, then the resulting class of "interval-rational" functions practically coincides with the class of all algebraic functions. By "practically coincides", we mean that first, every interval-rational function is algebraic, and that second, for every algebraic function A(x), there exists an interval-rational function that coincides with A(x) for almost all x. In [1], "almost all" was understood in terms of Lebesgue measure; this result is therefore, not very computer realistic, because all the numbers representable in a computer are rational and therefore, form a set of Lebesgue measure 0. In this article, we formulate a more computer-realistic version of the result that interval + rational = algebraic. |
---|---|
ISSN: | 0163-5778 |
DOI: | 10.1145/232794.232798 |