Source author record

Artūras Dubickas

Explore this author’s publications, collaborators and research topics. This catalog record has not been claimed by its author.

ResearcherUnclaimed source record

Catalog footprint

Research overview

5works
1topics
3close collaborators

Published work

5 publications

preprint2015arXiv

Counting and testing dominant polynomials

In this paper, we concentrate on counting and testing dominant polynomials with integer coefficients. A polynomial is called dominant if it has a simple root whose modulus is strictly greater than the moduli of its remaining roots. In particular, our results imply that the probability that the dominant root assumption holds for a random monic polynomial with integer coefficients tends to 1 in some setting. However, for arbitrary integer polynomials it does not tend to 1. For instance, the proportion of dominant quadratic integer polynomials of height HH among all quadratic integer polynomials tends to (41+6log⁡2)/72(41+6 \log 2)/72 as H→∞H \to \infty. Finally, we will design some algorithms to test whether a given polynomial with integer coefficients is dominant or not without finding the polynomial roots.

preprint2015arXiv

Counting degenerate polynomials of fixed degree and bounded height

In this paper, we give sharp upper and lower bounds for the number of degenerate monic (and arbitrary, not necessarily monic) polynomials with integer coefficients of fixed degree n≥2n \ge 2 and height bounded by H≥2H \ge 2. The polynomial is called degenerate if it has two distinct roots whose quotient is a root of unity. In particular, our bounds imply that non-degenerate linear recurrence sequences can be generated randomly.

preprint2015arXiv

Positive density of integer polynomials with some prescribed properties

In this paper, we show that various kinds of integer polynomials with prescribed properties of their roots have positive density. For example, we prove that almost all integer polynomials have exactly one or two roots with maximal modulus. We also show that for any positive integer nn and any set of nn distinct points symmetric with respect to the real line, there is a positive density of integer polynomials of degree nn, height at most HH and Galois group SnS_n whose roots are close to the given nn points.

preprint2014arXiv

There are no two non-real conjugates of a Pisot number with the same imaginary part

We show that the number α=(1+3+25)/2α=(1+\sqrt{3+2\sqrt{5}})/2 with minimal polynomial x4−2x3+x−1x^4-2x^3+x-1 is the only Pisot number whose four distinct conjugates α1,α2,α3,α4α_1,α_2,α_3,α_4 satisfy the additive relation α1+α2=α3+α4α_1+α_2=α_3+α_4. This implies that there exists no two non-real conjugates of a Pisot number with the same imaginary part and also that at most two conjugates of a Pisot number can have the same real part. On the other hand, we prove that similar four term equations α1=α2+α3+α4α_1 = α_2 + α_3+α_4 or α1+α2+α3+α4=0α_1 + α_2 + α_3 + α_4 =0 cannot be solved in conjugates of a Pisot number αα. We also show that the roots of the Siegel's polynomial x3−x−1x^3-x-1 are the only solutions to the three term equation α1+α2+α3=0α_1+α_2+α_3=0 in conjugates of a Pisot number. Finally, we prove that there exists no Pisot number whose conjugates satisfy the relation α1=α2+α3α_1=α_2+α_3.

Research graph

See the researcher in context

Open full explorer

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.