Abstract
The article gives an overview of power function variants and compares three differently extensive versions for use in interval arithmetics. Aiming at the general power function for inclusion in interval libraries, which is defined for as many pairs of base and exponent as possible, we observe that the definition of a power for each such pair depends on the context. This problem eventually comes up at the point where reasonable doubts about the definition \(0^0\) and powers with negative base in correlation with a non-integral exponent arise. We come up with several variants serving distinct purposes, yet also recommend a general-purpose power function, which is unique for being restricted to integral exponents on the domain of negative bases. Three different variants of interval power functions satisfy all meaningful general exponentiation needs. We provide a unified treatment to handle the variants and present valuable implementation techniques for the computation of these interval functions along with a mathematic foundation.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Private communication with Dan Zuras.
References
Averbukh B, Günther H (2008) On powers and power functions (Über die Potenzen und die Potenzfunktionen). Mathematikinformation (49):5–23 (german)
Chiriaev D, Walster GW (1998) Interval arithmetic specification. Tech. rep.
Daramy-Loirat C, Defour D, de Dinechin F, Gallet M, Gast N, Lauter CQ, Muller JM (2009) CR-LIBM a library of correctly rounded elementary functions on double-precision. http://lipforge.ens-lyon.fr/frs/download.php/153/crlibm-1.0beta3.pdf. Accessed Mar 25, 2012
IEEE 754, (2008) IEEE 754-2008, Standard for floating-point arithmetic. IEEE, New York
IEEE P1788 (2009) IEEE interval standard working group-P1788. http://grouper.ieee.org/groups/1788/
ISO C90 (1990) The ANSI C standard (C90). Tech. Rep. 9899:1990, ISO/IEC
Java 6 (2011) Java Platform, Standard Edition 6: API Specification. Oracle
Knuth D (1992) Two notes on notation. Am Math Mon 99(5):403–422
Krämer W, Wolff von Gudenberg J (2003) Extended interval power function. Relia Comput 9:339–347. doi:10.1023/A:1025175029490
Krämer W, Kulisch U, Lohner R (1994) Numerical toolbox for verified computing II: advanced numerical problems, Draft version. http://www2.math.uni-wuppertal.de/~xsc/literatur/tb2.pdf, Accessed Dec 13, 2011
Lauter CQ, Lefèvre V (2009) An efficient rounding boundary test for pow(x, y) in double precision. IEEE Trans Comput 58(2):197–207. doi:10.1109/TC.2008.202
Moebius A (1834) ‘Beweis der Gleichung \(0^{0}=1\) nach J. F. Pfaff’. Journal für die reine und angewandte Mathematik 12:134–136 (german)
Muller JM, Brisebarre N, de Dinecin F, Jeannerod CP, Lefèvre V, Melquion G, Revol N, Stehlè D, Torres S (2010) Handbook of floating point arithmetic. Birkhäuser, Berlin
Petrov E (2007) Algorithm for evaluation of the interval power function of unconstrained arguments. CoRR abs/0704.3141
Pryce JD, Corliss GF (2006) Interval arithmetic with containment sets. Computing 78(3):251–276. doi:10.1007/s00607-006-0180-4
Rump S (1999) INTLAB—INTerval LABoratory. In: Csendes T (ed) Developments in reliable computing. Kluwer Academic Publishers, Dordrecht, pp 77–104
Truss JK (1997) Foundations of mathematical analysis. Oxford Science Publications, Clarendon Press, Oxford
Wolff von Gudenberg J (2009) Motion 10, a Proposal for Elementary Functions. In: IEEE P1788(2009)
Acknowledgments
We thank the anonymous referees for their detailed and helpful comments.
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by V. Kreinovich.
Rights and permissions
About this article
Cite this article
Heimlich, O., Nehmeier, M. & Wolff von Gudenberg, J. Variants of the general interval power function. Soft Comput 17, 1357–1366 (2013). https://doi.org/10.1007/s00500-013-1008-8
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00500-013-1008-8