Correct rounding of algebraic functions

Jean-Michel Muller (1) , Nicolas Brisebarre (1, 2)

1 LIP - Laboratoire de l'Informatique du Parallélisme
2 LArAl - Laboratoire d'Arithmétique et d'Algèbre

We explicit the link between the computer arithmetic problem of providing correctly rounded algebraic functions and some diophantine approximation issues. This allows to get bounds on the accuracy with which intermediate calculations must be performed to correctly round these functions.

Keywords: Floating-point arithmetic; computer arithmetic; algebraic functions; correct rounding; diophantine approximation

Domains: Computer Science [cs]; Other [cs.OH]

Jean-Michel Muller, Nicolas Brisebarre. Correct rounding of algebraic functions. RAIRO - Theoretical Informatics and Applications (RAIRO: ITA), 2007, 41 (1), pp.71-83. ⟨10.1051/ita:2007002⟩. ⟨ensl-00143230⟩