Certification of a Numerical Result: Use of Interval Arithmetic and Multiple Precision

Hong Diep Nguyen 1 Nathalie Revol 1
1 ARENAIRE - Computer arithmetic
Inria Grenoble - Rhône-Alpes, LIP - Laboratoire de l'Informatique du Parallélisme
Abstract : Using floating-point arithmetic to solve a numerical problem yields a computed result, which is an approximation of the exact solution because of roundoff errors. In this paper, we present an approach to certify the computed solution. Here, "certify" means computing a guaranteed enclosure of the error between the computed, approximate, result and the exact, unknown result. We discuss an iterative refinement method: classically, such methods aim at computing an approximation of the error and they add it to the previous result to improve its accuracy. We add two ingredients: interval arithmetic is used to get an enclosure of the error instead of an approximation, and multiple precision is used to reach higher accuracy. We exemplify this approach on the certification of the solution of a linear system.
Document type :
Conference papers
Complete list of metadatas

Cited literature [15 references]  Display  Hide  Download

https://hal.inria.fr/inria-00544798
Contributor : Nathalie Revol <>
Submitted on : Thursday, December 9, 2010 - 12:55:55 AM
Last modification on : Thursday, February 7, 2019 - 3:45:25 PM
Long-term archiving on : Monday, November 5, 2012 - 12:50:32 PM

File

Nguyen-Revol-v2.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : inria-00544798, version 1

Collections

Citation

Hong Diep Nguyen, Nathalie Revol. Certification of a Numerical Result: Use of Interval Arithmetic and Multiple Precision. NSV-3: Third International Workshop on Numerical Software Verification., Fainekos, Georgios and Goubault, Eric and Putot, Sylvie, Jul 2010, Edinburgh, United Kingdom. ⟨inria-00544798⟩

Share

Metrics

Record views

200

Files downloads

129