A decision procedure for linear ``big O equations

From MaRDI portal
Publication:2642465

DOI10.1007/S10817-007-9066-1zbMATH Open1122.03004arXivcs/0701073OpenAlexW2140507616MaRDI QIDQ2642465FDOQ2642465


Authors: Jeremy Avigad, Kevin Donnelly Edit this on Wikidata


Publication date: 17 August 2007

Published in: Journal of Automated Reasoning (Search for Journal in Brave)

Abstract: Let F be the set of functions from an infinite set, S, to an ordered ring, R. For f, g, and h in F, the assertion f=g+O(h) means that for some constant C, |f(x)g(x)|leqC|h(x)| for every x in S. Let L be the first-order language with variables ranging over such functions, symbols for 0,+,,min,max, and absolute value, and a ternary relation f=g+O(h). We show that the set of quantifier-free formulas in this language that are valid in the intended class of interpretations is decidable, and does not depend on the underlying set, S, or the ordered ring, R. If R is a subfield of the real numbers, we can add a constant 1 function, as well as multiplication by constants from any computable subfield. We obtain further decidability results for certain situations in which one adds symbols denoting the elements of a fixed sequence of functions of strictly increasing rates of growth.


Full work available at URL: https://arxiv.org/abs/cs/0701073




Recommendations




Cites Work


Cited In (5)

Uses Software





This page was built for publication: A decision procedure for linear ``big O equations

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2642465)