Floating-point verification is a very interesting application area for theorem provers. HOL is a general-purpose prover which is equipped with an extensive and rigorous theory of real analysis. We explain how it can be used in floating point verification, illustrating our remarks with complete verifications of simple square-root and (natural) logarithm algorithms.
CITATION STYLE
Harrison, J. (1995). Floating point verification in HOL. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 971, pp. 186–199). Springer Verlag. https://doi.org/10.1007/3-540-60275-5_65
Mendeley helps you to discover research relevant for your work.