NASA Logo

NTRS

NTRS - NASA Technical Reports Server

Back to Results
Stochastic Formal Correctness of Numerical AlgorithmsWe provide a framework to bound the probability that accumulated errors were never above a given threshold on numerical algorithms. Such algorithms are used for example in aircraft and nuclear power plants. This report contains simple formulas based on Levy's and Markov's inequalities and it presents a formal theory of random variables with a special focus on producing concrete results. We selected four very common applications that fit in our framework and cover the common practices of systems that evolve for a long time. We compute the number of bits that remain continuously significant in the first two applications with a probability of failure around one out of a billion, where worst case analysis considers that no significant bit remains. We are using PVS as such formal tools force explicit statement of all hypotheses and prevent incorrect uses of theorems.
Document ID
20100024459
Acquisition Source
Ames Research Center
Document Type
Conference Paper
Authors
Daumas, Marc
(Perpignan Univ. Perpignan, France)
Lester, David
(Perpignan Univ. Perpignan, France)
Martin-Dorel, Erik
(Perpignan Univ. Perpignan, France)
Truffert, Annick
(Manchester Univ. United Kingdom)
Date Acquired
August 24, 2013
Publication Date
April 1, 2009
Publication Information
Publication: Proceedings of the First NASA Formal Methods Symposium
Subject Category
Mathematical And Computer Sciences (General)
Funding Number(s)
CONTRACT_GRANT: CNRS PICS 2533
Distribution Limits
Public
Copyright
Public Use Permitted.
No Preview Available