2006•Unpublished venueRequires access

An Optimal Bound for Sum of Square Roots of Special Type of Integers

Jianbo Qian, Cao An Wang

Open publisher page 0 citations

Abstract

Abstract The sum of square roots of integers problem is to find the minimum nonzero difference between two sums of square roots of integers. Let r(n,k) denote the minimum nonzero positive value: | ∑ k √ i=1 ai − ∑ k √ i=1 bi|, where ai and bi are positive integers not larger than integer n. We prove by an explicit construction that r(n,k) = O(n−2k+ 3 2) for fixed k and any n. Our result implies that in order to compare two sums of k square roots of integers with at most d digits per integer, one might need precision of as many as (2k − 3)d digits. We also prove that this bound is optimal 2 for a wide range of integers, i.e., r(n,k) = Θ(n−2k+ 3 2) for fixed k and for those integers in the form of n = � �2 � �2 2k−1

About this research paper

What this paper is about

Abstract The sum of square roots of integers problem is to find the minimum nonzero difference between two sums of square roots of integers. Let r(n,k) denote the minimum nonzero positive value: | ∑ k √ i=1 ai − ∑ k √ i=1 bi|, where ai and bi are positive integers not larger than integer n. We prove by an explicit construction that r(n,k) = O(n−2k+ 3 2) for fixed k and any n. Our result implies that in order to compare two sums of k square roots of integers with at most d digits per integer, one might need precision of as many as (2k − 3)d digits. We also prove that this bound is optimal 2 for a wide range of integers, i.e., r(n,k) = Θ(n−2k+ 3 2) for fixed k and for those integers in the form of n = � �2 � �2 2k−1

Why it matters

A significance statement is not available in the OpenAlex record.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Abstract The sum of square roots of integers problem is to find the minimum nonzero difference between two sums of square roots of integers. Let r(n,k) denote the minimum nonzero positive value: | ∑ k √ i=1 ai − ∑ k √ i=1 bi|, where ai and bi are positive integers not larger than integer n. We prove by an explicit construction that r(n,k) = O(n−2k+ 3 2) for fixed k and any n. Our result implies that in order to compare two sums of k square roots of integers with at most d digits per integer, one might need precision of as many as (2k − 3)d digits. We also prove that this bound is optimal 2 for a wide range of integers, i.e., r(n,k) = Θ(n−2k+ 3 2) for fixed k and for those integers in the form of n = � �2 � �2 2k−1

Key concepts: Mathematics, Integer (computer science), Combinatorics, Square (algebra), Square number, Order (exchange), Upper and lower bounds, Discrete mathematics

Related papers

Back to paper searchBrowse research topicsOriginal source
An Optimal Bound for Sum of Square Roots of Special Type of Integers — Research Paper | ScholarLens