Fast evaluation and root finding for polynomials with floating-point coefficients
Résumé
Evaluating or finding the roots of a polynomial f (z) = f 0 + • • • + f d z d with floating-point number coefficients is a ubiquitous problem. By using a piecewise approximation of f obtained with a careful use of the Newton polygon of f , we improve state-of-the-art upper bounds on the number of operations to evaluate and find the roots of a polynomial. In particular, if the coefficients of f are given with m significant bits, we provide for the first time an algorithm that finds all the roots of f with a relative condition number lower than 2 m , using a number of bit operations quasi-linear in the bit-size of the floating-point representation of f . Notably, our new approach handles efficiently polynomials with coefficients ranging from 2 -d to 2 d , both in theory and in practice.
Fichier principal
jsc-issac-2023_imbach-moroz.pdf (1.08 Mo)
Télécharger le fichier
sectors_ex.png (90.49 Ko)
Télécharger le fichier
Origine | Fichiers produits par l'(les) auteur(s) |
---|
Format | Figure, Image |
---|