Skip to main content
Log in

Improved bound for stochastic formal correctness of numerical algorithms

  • Original Paper
  • Published:
Innovations in Systems and Software Engineering Aims and scope Submit manuscript

Abstract

We provide bounds on 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 Lévy’s, Markov’s and Hoeffding’s inequalities and it presents a formal theory of random variables with a special focus on producing concrete results. We select three very common applications that 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.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  1. Audebaud P, Paulin-Mohring C (2006) Proofs of randomized algorithms in Coq. In: Uustalu T (ed) Proceedings of the 8th international conference on mathematics of program construction. Kuressaare, Estonia, pp 49–68. doi:10.1007/11783596_6

  2. Bertoin J (2001) Probabilités. http://www.proba.jussieu.fr/cours/bertoin.pdf. Cours de licence de mathématiques appliquées

  3. Boldo S, Daumas M (2003) Representable correcting terms for possibly underflowing floating point operations. In: Bajard JC, Schulte M (eds) Proceedings of the 16th symposium on computer arithmetic. Santiago de Compostela, Spain, pp 79–86. http://perso.ens-lyon.fr/marc.daumas/SoftArith/BolDau03.pdf

  4. Boldo S, Muñoz C (2006) Provably faithful evaluation of polynomials. In: Proceedings of the 2006 ACM symposium on applied computing. Dijon, France, pp 1328–1332. doi:10.1145/1141277.1141586

  5. Bustoz J, Feldstein A, Goodman R, Linnainmaa S (1979) Improved trailing digits estimates applied to optimal computer arithmetic. J ACM 26(4):716–730. doi:10.1145/322154.322162

    Article  MATH  MathSciNet  Google Scholar 

  6. Chesneaux JM (1990) Contribution to computer arithmetic and self-validating numerical methods. In: Ullrich C (eds) Study of the computing accuracy by using probabilistic approach. Baltzer, Basel, pp 19–30

    Google Scholar 

  7. Daumas M, Lester D (2007) Stochastic formal methods: an application to accuracy of numeric software. In: Proceedings of the 40th IEEE annual Hawaii international conference on system sciences, p 7. Waikoloa, Hawaii. http://hal.ccsd.cnrs.fr/ccsd-00081413

  8. Daumas M, Lester D, Martin-Dorel É, Truffert A (2009) Stochastic formal correctness of numerical algorithms. In: NASA formal methods symposium, pp 136–145. http://ti.arc.nasa.gov/m/event/nfm09/NFM09Proceedings.pdf

  9. Daumas M, Lester D, Muñoz C (2009) Verified real number calculations: a library for interval arithmetic. IEEE Trans Comput 58(2): 226–237. doi:10.1109/TC.2008.213

    Article  Google Scholar 

  10. Daumas M, Melquiond G (2010) Certification of bounds on expressions involving rounded operators. ACM Trans Math Softw 37(1). http://hal.archives-ouvertes.fr/hal-00127769 (to appear)

  11. Feldstein A, Goodman R (1976) Convergence estimates for the distribution of trailing digits. J ACM 23(2): 287–297. doi:10.1145/321941.321948

    Article  MATH  MathSciNet  Google Scholar 

  12. Goldberg D (1991) What every computer scientist should know about floating point arithmetic. ACM Comput Surv 23(1): 5–47. doi:10.1145/103162.103163

    Article  Google Scholar 

  13. Gordon MJC, Melham TF (eds) (1993) Introduction to HOL: A theorem proving environment for higher order logic. Cambridge University Press, Cambridge

    MATH  Google Scholar 

  14. Harrison J (2000) Formal verification of floating point trigonometric functions. In: Hunt WA, Johnson SD (eds) Proceedings of the third international conference on formal methods in computer-aided design, pp 217–233. Austin, Texas. http://www.springerlink.com/link.asp?id=wxvaqu9wjrgc8l99

  15. Huet G, Kahn G, Paulin-Mohring C (2009) The Coq proof assistant: a tutorial: version 8.2. http://coq.inria.fr/distrib/current/files/Tutorial.pdf

  16. Hurd J (2002) Formal verification of probabilistic algorithms. Ph.D. thesis, University of Cambridge. http://www.cl.cam.ac.uk/~jeh1004/research/papers/thesis.pdf

  17. Kaufmann M, Manolios P, Moore JS (2000) Computer-aided reasoning: an approach. Kluwer, Dordrecht

    Google Scholar 

  18. Knuth DE (1997) The art of computer programming: seminumerical algorithms, 3rd edn. Addison-Wesley, Reading

    Google Scholar 

  19. Neveu J (ed) (1972) Martingales à temps discret. Masson, Paris

    Google Scholar 

  20. Owre S, Rushby JM, Shankar N (1992) PVS: a prototype verification system. In: Kapur D (ed) 11th international conference on automated deduction. Springer, Saratoga, New York, pp 748–752. http://pvs.csl.sri.com/papers/cade92-pvs/cade92-pvs.ps

  21. Russinoff DM (1998) A mechanically checked proof of IEEE compliance of the floating point multiplication, division and square root algorithms of the AMD-K7 processor. LMS J Comput Math 1:148–200. http://www.onr.com/user/russ/david/k7-div-sqrt.ps

    Google Scholar 

  22. Stevenson D et al (1987) An American national standard: IEEE standard for binary floating point arithmetic. ACM SIGPLAN Notices 22(2): 9–25

    Google Scholar 

  23. Texas Instruments (1997) TMS320C3x—user’s guide. http://www.s.ti.com/sc/psheets/spru031e/spru031e.pdf

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Érik Martin-Dorel.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Daumas, M., Lester, D., Martin-Dorel, É. et al. Improved bound for stochastic formal correctness of numerical algorithms. Innovations Syst Softw Eng 6, 173–179 (2010). https://doi.org/10.1007/s11334-010-0128-x

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11334-010-0128-x

Keywords

Navigation