Source author record

Maciej Ulas

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

24works
4topics
4close collaborators

Published work

24 publications

preprint2022arXiv

On the Diophantine equation σ2(X‾n)=σn(X‾n)σ_{2}(\overline{X}_{n})=σ_{n}(\overline{X}_{n})

In this note we investigate the set S(n)S(n) of positive integer solutions of the title Diophantine equation. In particular, for a given nn we prove boundedness of the number of solutions, give precise upper bound on the common value of σ2(X‾n)σ_{2}(\overline{X}_{n}) and σn(X‾n)σ_{n}(\overline{X}_{n}) together with the biggest value of the variable xnx_{n} appearing in the solution. Moreover, we enumerate all solutions for n≤16n\leq 16 and discuss the set of values of xn/xn−1x_{n}/x_{n-1} over elements of S(n)S(n).

preprint2022arXiv

Signs behaviour of sums of weighted numbers of partitions

Let AA be a subset of positive integers. By AA-partition of nn we understand the representation of nn as a sum of elements from the set AA. For given i,n∈Ni, n\in\N, by cA(i,n)c_{A}(i,n) we denote the number of AA-partitions of nn with exactly ii parts. In the paper we obtain several result concerning sign behaviour of the sequence SA,k(n)=∑i=0n(−1)iikcA(i,n)S_{A,k}(n)=\sum_{i=0}^{n}(-1)^{i}i^{k}c_{A}(i,n), where k∈Nk\in\N is fixed. In particular, we prove that for a broad class A\cal{A} of subsets of N+\N_{+} we have that for each A∈AA\in \cal{A} we have (−1)nSA,k(n)≥0(-1)^{n}S_{A,k}(n)\geq 0 for each n,k∈Nn, k\in\N.

preprint2021arXiv

Some observations and speculations on partitions into dd-th powers

The aim of this note is to provoke discussion concerning arithmetic properties of function pd(n)p_{d}(n) counting partitions of an positive integer nn into dd-th powers, where d≥2d\geq 2. Besides results concerning the asymptotic behavior of pd(n)p_{d}(n) a little is known. In the first part of the paper, we prove certain congruences involving functions counting various types of partitions into dd-th powers. The second part of the paper has experimental nature and contains questions and conjectures concerning arithmetic behavior of the sequence (pd(n))n∈N(p_{d}(n))_{n\in\N}. They based on our computations of pd(n)p_{d}(n) for n≤105n\leq 10^5 in case of d=2d=2, and n≤106n\leq 10^{6} for d=3,4,5d=3, 4, 5.

preprint2020arXiv

On a Diophantine equation of Erdős and Graham

We study solvability of the Diophantine equation \begin{equation*} \frac{n}{2^{n}}=\sum_{i=1}^{k}\frac{a_{i}}{2^{a_{i}}}, \end{equation*} in integers n,k,a1,…,akn, k, a_{1},\ldots, a_{k} satisfying the conditions k≥2k\geq 2 and ai<ai+1a_{i}<a_{i+1} for i=1,…,k−1i=1,\ldots,k-1. The above Diophantine equation (of polynomial-exponential type) was mentioned in the monograph of Erdős and Graham, where several questions were stated. Some of these questions were already answered by Borwein and Loring. We extend their work and investigate other aspects of Erdős and Graham equation. First of all, we obtain the upper bound for the value aka_{k} given in terms of kk only. This mean, that with fixed kk our equation has only finitely many solutions in n,a1,…,akn, a_{1},\ldots, a_{k}. Moreover, we construct an infinite set K\cal{K}, such that for each k∈Kk\in\cal{K}, the considered equation has at least five solutions. As an application of our findings we enumerate all solutions of the equation for k≤8k\leq 8. Moreover, by applying greedy algorithm, we extend Borwein and Loring calculations and check that for each n≤104n\leq 10^4 there is a value of kk such that the considered equation has a solution in integers n+1=a1<a2<…<akn+1=a_{1}<a_{2}<\ldots <a_{k}. Based on our numerical calculations we formulate some further questions and conjectures.

preprint2020arXiv

Some Properties of a Class of Sparse Polynomials

We study an infinite class of sequences of sparse polynomials that have binomial coefficients both as exponents and as coefficients. This generalizes a sequence of sparse polynomials which arises in a natural way as graph theoretic polynomials. After deriving some basic identities, we obtain properties concerning monotonicity and log-concavity, as well as identities involving derivatives. We also prove upper and lower bounds on the moduli of the zeros of these polynomials.

preprint2016arXiv

On formal inverse of the Prouhet-Thue-Morse sequence

Let pp be a prime number and consider a pp-automatic sequence u=(un)n∈N{\bf u}=(u_{n})_{n\in\N} and its generating function U(X)=∑n=0∞unXn∈Fp[[X]]U(X)=\sum_{n=0}^{\infty}u_{n}X^{n}\in\mathbb{F}_{p}[[X]]. Moreover, let us suppose that u0=0u_{0}=0 and u1≠0u_{1}\neq 0 and consider the formal power series V∈Fp[[X]]V\in\mathbb{F}_{p}[[X]] which is a compositional inverse of U(X)U(X), i.e., U(V(X))=V(U(X))=XU(V(X))=V(U(X))=X. In this note we initiate the study of arithmetic properties of the sequence of coefficients of the power series V(X)V(X). We are mainly interested in the case when un=tnu_{n}=t_{n}, where tn=s2(n)(mod2)t_{n}=s_{2}(n)\pmod{2} and t=(tn)n∈N{\bf t}=(t_{n})_{n\in\N} is the Prouhet-Thue-Morse sequence defined on the two letter alphabet {0,1}\{0,1\}. More precisely, we study the sequence c=(cn)n∈N{\bf c}=(c_{n})_{n\in\N} which is the sequence of coefficients of the compositional inverse of the generating function of the sequence t{\bf t}. This sequence is clearly 2-automatic. We describe the sequence a{\bf a} characterizing solutions of the equation cn=1c_{n}=1. In particular, we prove that the sequence a{\bf a} is 2-regular. We also prove that an increasing sequence characterizing solutions of the equation cn=0c_{n}=0 is not kk-regular for any kk. Moreover, we present a result concerning some density properties of a sequence related to a{\bf a}.

preprint2015arXiv

On primitive integer solutions of the Diophantine equation t2=G(x,y,z)t^2=G(x,y,z) and related results

In this paper we investigate Diophantine equations of the form T2=G(X‾),  X‾=(X1,…,Xm)T^2=G(\overline{X}),\; \overline{X}=(X_{1},\ldots,X_{m}), where m=3m=3 or m=4m=4 and GG is specific homogenous quintic form. First, we prove that if F(x,y,z)=x2+y2+az2+bxy+cyz+dxz∈Z[x,y,z]F(x,y,z)=x^2+y^2+az^2+bxy+cyz+dxz\in\Z[x,y,z] and (b−2,4a−d2,d)≠(0,0,0)(b-2,4a-d^2,d)\neq (0,0,0), then the Diophantine equation t2=nxyzF(x,y,z)t^2=nxyzF(x,y,z) has solution in polynomials x,y,z,tx, y, z, t with integer coefficients, without polynomial common factor of positive degree. In case a=d=0,b=2a=d=0, b=2 we prove that there are infinitely many primitive integer solutions of the Diophantine equation under consideration. As an application of our result we prove that for each $n\in\Q\setminus\{0\}$ the Diophantine equation \begin{equation*} T^2=n(X_{1}^5+X_{2}^5+X_{3}^5+X_{4}^5) \end{equation*} has a solution in co-prime (non-homogenous) polynomials in two variables with integer coefficients. We also present a method which sometimes allow us to prove the existence of primitive integers solutions of more general quintic Diophantine equation of the form T2=aX15+bX25+cX35+dX45T^2=aX_{1}^5+bX_{2}^5+cX_{3}^5+dX_{4}^5, where a,b,c,d∈Za, b, c, d\in\Z. In particular, we prove that for each m,n∈Z∖{0},m, n\in\Z\setminus\{0\}, the Diophantine equation \begin{equation*} T^2=m(X_{1}^5-X_{2}^5)+n^2(X_{3}^5-X_{4}^5) \end{equation*} has a solution in polynomials which are co-prime over Z[t]\Z[t]. Moreover, we show how modification of the presented method can be used in order to prove that for each $n\in\Q\setminus\{0\}$, the Diophantine equation \begin{equation*} t^2=n(X_{1}^5+X_{2}^5-2X_{3}^5) \end{equation*} has a solution in polynomials which are co-prime over Z[t]\Z[t].

preprint2014arXiv

Constructions of diagonal quartic and sextic surfaces with infinitely many rational points

In this note we construct several infinite families of diagonal quartic surfaces \begin{equation*} ax^4+by^4+cz^4+dw^4=0, \end{equation*} where a,b,c,d∈Z∖{0}a,b,c,d\in\Z\setminus\{0\} with infinitely many rational points and satisfying the condition abcd≠□abcd\neq \square. In particular, we present an infinite family of diagonal quartic surfaces defined over $\Q$ with Picard number equal to one and possessing infinitely many rational points. Further, we present some sextic surfaces of type ax6+by6+cz6+dwi=0ax^6+by^6+cz^6+dw^i=0, i=2i=2, 33, or 66, with infinitely many rational points.

preprint2014arXiv

Some experiments with Ramanujan-Nagell type Diophantine equations

Stiller proved that the Diophantine equation x2+119=15⋅2nx^2+119=15\cdot 2^{n} has exactly six solutions in positive integers. Motivated by this result we are interested in constructions of Diophantine equations of Ramanujan-Nagell type x2=Akn+Bx^2=Ak^{n}+B with many solutions. Here, A,B∈ZA,B\in\Z (thus A,BA, B are not necessarily positive) and k∈Z≥2k\in\Z_{\geq 2} are given integers. In particular, we prove that for each kk there exists an infinite set S\cal{S} containing pairs of integers (A,B)(A, B) such that for each (A,B)∈S(A,B)\in \cal{S} we have gcd⁡(A,B)\gcd(A,B) is square-free and the Diophantine equation x2=Akn+Bx^2=Ak^n+B has at least four solutions in positive integers. Moreover, we construct several Diophantine equations of the form x2=Akn+Bx^2=Ak^n+B with k>2k>2, each containing five solutions in non-negative integers. %For example the equation y2=130⋅3n+5550606y^2=130\cdot 3^{n}+5550606 has exactly five solutions with n=0,6,11,15,16n=0, 6, 11, 15, 16. We also find new examples of equations x2=A2n+Bx^2=A2^{n}+B having six solutions in positive integers, e.g. the following Diophantine equations has exactly six solutions: \begin{equation*} \begin{array}{ll} x^2= 57\cdot 2^{n}+117440512 & n=0, 14, 16, 20, 24, 25, x^2= 165\cdot 2^{n}+26404 & n=0, 5, 7, 8, 10, 12. \end{array} \end{equation*} Moreover, based on an extensive numerical calculations we state several conjectures on the number of solutions of certain parametric families of the Diophantine equations of Ramanujan-Nagell type.

preprint2014arXiv

Some identities involving Prouhet-Thue-Morse sequence and its relatives

Let sk(n)s_{k}(n) denote the sum of digits of an integer nn in base kk. Motivated by certain identities of Nieto, and Bateman and Bradley involving sums of the form ∑i=02n−1(−1)s2(i)(x+i)m\sum_{i=0}^{2^{n}-1}(-1)^{s_{2}(i)}(x+i)^{m} for m=nm=n and m=n+1m=n+1, we consider the sequence of polynomials \begin{equation*} f_{m,n}^{\mathbf u}(x)=\sum_{i=0}^{k^{n}-1}ζ_{k}^{s_{k}(i)}(x+{\mathbf u}(i))^{m}. \end{equation*} defined for sequences u(i){\bf u}(i) satisfying a certain recurrence relation. We prove that computing these polynomials is essentially equivalent with computing their constant term and we find an explicit formula for this number. This allows us to prove several interesting identities involving sums of binary digits. We also prove some related results which are of independent interests and can be seen as further generalizations of certain sums involving Prouhet-Thue-Morse sequence.

preprint2014arXiv

Variations on twists of tuples of hyperelliptic curves and related results

Let $f\in\Q[x]$ be a square-free polynomial of degree ≥3\geq 3 and m≥3m\geq 3 be an odd positive integer. Based on our earlier investigations we prove that there exists a function $D_{1}\in\Q(u,v,w)$ such that the Jacobians of the curves \begin{equation*} C_{1}:\;D_{1}y^2=f(x),\quad C_{2}:\;y^2=D_{1}x^m+b,\quad C_{3}:\;y^2=D_{1}x^m+c, \end{equation*} have all positive ranks over $\Q(u,v,w)$. Similarly, we prove that there exists a function $D_{2}\in\Q(u,v,w)$ such that the Jacobians of the curves \begin{equation*} C_{1}:\;D_{2}y^2=h(x),\quad C_{2}:\;y^2=D_{2}x^m+b,\quad C_{3}:\;y^2=x^m+cD_{2}, \end{equation*} have all positive ranks over $\Q(u,v,w)$. Moreover, if f(x)=xm+af(x)=x^m+a for some a∈Z∖{0}a\in\Z\setminus\{0\}, we prove the existence of a function $D_{3}\in\Q(u,v,w)$ such that the Jacobians of the curves \begin{equation*} C_{1}:\;y^2=D_{3}x^{m}+a,\quad C_{2}:\;y^2=D_{3}x^m+b,\quad C_{3}:\;y^2=x^m+cD_{3}, \end{equation*} have all positive ranks over $\Q(u,v,w)$. We present also some applications of these results. Finally, we present some results concerning the torsion parts of the Jacobians of the superelliptic curves yp=xm(x+a)y^p=x^{m}(x+a) and yp=xm(a−x)ky^p=x^{m}(a-x)^{k} for a prime pp and 0<m<p−20<m<p-2 and k<pk<p and apply our result in order to prove the existence of a function $D\in\Q(u,v,w,t)$ such that the Jacobians of the curves \begin{equation*} C_{1}:\;Dy^p=x^m(x+a),\quad Dy^p=x^m(x+b) \end{equation*} have both positive rank over $\Q(u,v,w,t)$.

preprint2013arXiv

A note on Diophantine systems involving three symmetric polynomials

Let Xˉn=(x1,…,xn)\bar{X}_{n}=(x_{1},\ldots,x_{n}) and σi(Xˉn)=∑xk1…xkiσ_{i}(\bar{X}_{n})=\sum x_{k_{1}}\ldots x_{k_{i}} be ii-th elementary symmetric polynomial. In this note we prove that there are infinitely many triples of integers a,b,ca, b, c such that for each 1≤i≤n1\leq i\leq n the system of Diophantine equations \begin{equation*} σ_{i}(\bar{X}_{2n})=a, \quad σ_{2n-i}(\bar{X}_{2n})=b, \quad σ_{2n}(\bar{X}_{2n})=c \end{equation*} has infinitely many rational solutions. This result extend the recent results of Zhang and Cai, and the author. Moreover, we also consider some Diophantine systems involving sums of powers. In particular, we prove that for each kk there are at least kk nn-tuples of integers with the same sum of ii-th powers for i=1,2,3i=1,2,3. Similar result is proved for i=1,2,4i=1,2,4 and i=−1,1,2i=-1,1,2.

preprint2013arXiv

On certain diophantine equations of diagonal type

In this note we consider Diophantine equations of the form \begin{equation*} a(x^p-y^q) = b(z^r-w^s), \quad \mbox{where}\quad \frac{1}{p}+\frac{1}{q}+\frac{1}{r}+\frac{1}{s}=1, \end{equation*} with even positive integers p,q,r,sp,q,r,s. We show that in each case the set of rational points on the underlying surface is dense in the Zariski topology. For the surface with (p,q,r,s)=(2,6,6,6)(p,q,r,s)=(2,6,6,6) we prove density of rational points in the Euclidean topology. Moreover, in this case we construct infinitely many parametric solutions in coprime polynomials. The same result is true for (p,q,r,s)∈{(2,4,8,8),(2,8,4,8)}(p,q,r,s)\in\{(2,4,8,8), (2,8,4,8)\}. In the case (p,q,r,s)=(4,4,4,4)(p,q,r,s)=(4,4,4,4), we present some new parametric solutions of the equation x4−y4=4(z4−w4)x^4-y^4=4(z^4-w^4).

preprint2013arXiv

On some Diophantine systems involving symmetric polynomials

Let σi(x1,…,xn)=∑1≤k1<k2<…<ki≤nxk1…xkiσ_{i}(x_{1},\ldots, x_{n})=\sum_{1\leq k_{1}<k_{2}<\ldots <k_{i}\leq n}x_{k_{1}}\ldots x_{k_{i}} be the ii-th elementary symmetric polynomial. In this note we generalize and extend the results obtained in a recent work of Zhang and Cai \cite{ZC,ZC2}. More precisely, we prove that for each n≥4n\geq 4 and rational numbers a,ba, b with ab≠0ab\neq 0, the system of diophantine equations \begin{equation*} σ_{1}(x_{1},\ldots, x_{n})=a, \quad σ_{n}(x_{1},\ldots, x_{n})=b, \end{equation*} has infinitely many solutions depending on n−3n-3 free parameters. A similar result is proved for the system \begin{equation*} σ_{i}(x_{1},\ldots, x_{n})=a, \quad σ_{n}(x_{1},\ldots, x_{n})=b, \end{equation*} with n≥4n\geq 4 and 2≤i<n2\leq i< n. Here, a,ba, b are rational numbers with b≠0b\neq 0. We also give some results concerning the general system of the form \begin{equation*} σ_{i}(x_{1},\ldots, x_{n})=a, \quad σ_{j}(x_{1},\ldots, x_{n})=b, \end{equation*} with suitably chosen rational values of a,ba, b and i<j<ni<j<n. Finally, we present some remarks on the systems involving three different symmetric polynomials.

preprint2013arXiv

Rational solutions of certain Diophantine equations involving norms

In this note we present some results concerning the unirationality of the algebraic variety Sf\cal{S}_{f} given by the equation \begin{equation*} N_{K/k}(X_{1}+αX_{2}+α^2 X_{3})=f(t), \end{equation*} where kk is a number field, K=k(α)K=k(α), αα is a root of an irreducible polynomial h(x)=x3+ax+b∈k[x]h(x)=x^3+ax+b\in k[x] and f∈k[t]f\in k[t]. We are mainly interested in the case of pure cubic extensions, i.e. a=0a=0 and b∈k∖k3b\in k\setminus k^{3}. We prove that if $\op{deg}f=4$ and the variety Sf\cal{S}_{f} contains a kk-rational point (x0,y0,z0,t0)(x_{0},y_{0},z_{0},t_{0}) with f(t0)≠0f(t_{0})\neq 0, then Sf\cal{S}_{f} is kk-unirational. A similar result is proved for a broad family of quintic polynomials ff satisfying some mild conditions (for example this family contains all irreducible polynomials). Moreover, the unirationality of Sf\cal{S}_{f} (with non-trivial kk-rational point) is proved for any polynomial ff of degree 6 with ff not equivalent to the polynomial hh satisfying the condition h(t)≠h(ζ3t)h(t)\neq h(ζ_{3}t), where ζ3ζ_{3} is the primitive third root of unity. We are able to prove the same result for an extension of degree 3 generated by the root of polynomial h(x)=x3+ax+b∈k[x]h(x)=x^3+ax+b\in k[x], provided that f(t)=t6+a4t4+a1t+a0∈k[t]f(t)=t^6+a_{4}t^4+a_{1}t+a_{0}\in k[t] with a1a4≠0a_{1}a_{4}\neq 0.

preprint2013arXiv

Some observations concerning reducibility of quadrinomials

In a recent paper \cite{Jan}, Jankauskas proved some interesting results concerning the reducibility of quadrinomials of the form f(4,x)f(4,x), where f(a,x)=xn+xm+xk+af(a,x)=x^{n}+x^{m}+x^{k}+a. He also obtained some examples of reducible quadrinomials f(a,x)f(a,x) with a∈Za\in\Z, such that all the irreducible factors of f(a,x)f(a,x) are of degree ≥3\geq 3. In this paper we perform a more systematic approach to the problem and ask about reducibility of f(a,x)f(a,x) with $a\in\Q$. In particular by computing the set of rational points on some genus two curves we characterize in several cases all quadrinomials f(a,x)f(a,x) with degree ≤6\leq 6 and divisible by a quadratic polynomial. We also give further examples of reducible f(a,x)f(a,x), $a\in\Q$, such that all irreducible factors are of degree ≥3\geq 3.

preprint2011arXiv

Arithmetic properties of the sequence of degrees of Stern polynomials and related results

Let Bn(t)B_{n}(t) be a nn-th Stern polynomial and let $e(n)=\op{deg}B_{n}(t)$ be its degree. In this note we continue our study started in \cite{Ul} of the arithmetic properties of the sequence of Stern polynomials and the sequence {e(n)}n=1∞\{e(n)\}_{n=1}^{\infty}. We also study the sequence $d(n)=\op{ord}_{t=0}B_{n}(t)$. Among other things we prove that d(n)=ν(n)d(n)=ν(n), where ν(n)ν(n) is the maximal power of 2 which dividies the number nn. We also count the number of the solutions of the equations e(m)=ie(m)=i and e(m)−d(m)=ie(m)-d(m)=i in the interval [1,2n][1,2^{n}]. We also obtain an interesting closed expression for a certain sum involving Stern polynomials.

preprint2011arXiv

On the reducibility type of trinomials

Say a trinomial $x^n+A x^m+B \in \Q[x]$ has reducibility type (n1,n2,...,nk)(n_1,n_2,...,n_k) if there exists a factorization of the trinomial into irreducible polynomials in $\Q[x]$ of degrees n1n_1, n2n_2,...,nkn_k, ordered so that n1≤n2≤...≤nkn_1 \leq n_2 \leq ... \leq n_k. Specifying the reducibility type of a monic polynomial of fixed degree is equivalent to specifying rational points on an algebraic curve. When the genus of this curve is 0 or 1, there is reasonable hope that all its rational points may be described; and techniques are available that may also find all points when the genus is 2. Thus all corresponding reducibility types may be described. These low genus instances are the ones studied in this paper.

preprint2008arXiv

Rational points on certain quintic hypersurfaces

Let f(x)=x5+ax3+bx2+cx∈Z[x]f(x)=x^5+ax^3+bx^2+cx \in \Z[x] and consider the hypersurface of degree five given by the equation \cal{V}_{f}: f(p)+f(q)=f(r)+f(s). Under the assumption b≠0b\neq 0 we show that there exists $\Q$-unirational elliptic surface contained in Vf\cal{V}_{f}. If b=0,a<0b=0, a<0 and −a≢2,18,34(mod48)-a\not\equiv 2,18,34 \pmod {48} then there exists $\Q$-rational surface contained in Vf\cal{V}_{f}. Moreover, we prove that for each ff of degree five there exists $\Q(i)$-rational surface contained in Vf\cal{V}_{f}.

preprint2007arXiv

Rational points on certain elliptic surfaces

Let Ef:y2=x3+f(t)x\mathcal{E}_{f}:y^2=x^3+f(t)x, where $f\in\Q[t]\setminus\Q$, and let us assume that $\op{deg}f\leq 4$. In this paper we prove that if $\op{deg}f\leq 3$, then there exists a rational base change t↦ϕ(t)t\mapstoϕ(t) such that on the surface Ef∘ϕ\cal{E}_{f\circϕ} there is a non-torsion section. A similar theorem is valid in case when $\op{deg}f=4$ and there exists $t_{0}\in\Q$ such that infinitely many rational points lie on the curve Et0:y2=x3+f(t0)xE_{t_{0}}:y^2=x^3+f(t_{0})x. In particular, we prove that if $\op{deg}f=4$ and ff is not an even polynomial, then there is a rational point on Ef\cal{E}_{f}. Next, we consider a surface Eg:y2=x3+g(t)\cal{E}^{g}:y^2=x^3+g(t), where $g\in\Q[t]$ is a monic polynomial of degree six. We prove that if the polynomial gg is not even, there is a rational base change t↦ψ(t)t\mapstoψ(t) such that on the surface Eg∘ψ\cal{E}^{g\circψ} there is a non-torsion section. Furthermore, if there exists $t_{0}\in\Q$ such that on the curve Et0:y2=x3+g(t0)E^{t_{0}}:y^2=x^3+g(t_{0}) there are infinitely many rational points, then the set of these t0t_{0} is infinite. We also present some results concerning diophantine equation of the form x2−y3−g(z)=tx^2-y^3-g(z)=t, where tt is a variable.

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.