SearcharxivSearch

arXiv subjects

Aliaksei Semchankau

Publications and source records attributed to Aliaksei Semchankau.

6 recordsLinked to original sources

New Upper bounds on the Mondrian Art Problem

We present a new upper bound on the defect of the Mondrian Art Problem. The Mondrian Art Problem asks for a partition of an $n \times n$ square with rectangles of distinct dimensions such that the difference (defect) between the largest and smallest rectangle areas is minimized. We prove that for any $n \times n$ square, there exists a partition with defect $O(n^{5/6})$, improving upon the previously conjectured $O (n/\log n)$ upper bound. We also implement an algorithm that provides empirical evidence supporting our theoretical bound.

math.CO

Structure theory of set addition with two operations

We take the first step toward a structure theory that includes both operations of a ring $\mathcal{R}$. More precisely, we prove a series of inverse results for the structure of sets $A\subseteq \mathbf{F}_p$ such that, under certain conditions on integers $r_1, \dots, r_k$, one has $|A^{r_1} + \dots + A^{r_k}| \ll \sqrt[k]{p^{k-1} |A|}$.

math.CO

A new bound for $A(A + A)$ for large sets

For $p$ being a large prime number, and $A \subset \mathbb{F}_p$ we prove the following: $(i)$ If $A(A+A)$ does not cover all nonzero residues in $\mathbb{F}_p$, then $|A| < p/8 + o(p)$. $(ii)$ If $A$ is both sum-free and satisfies $A = A^*$, then $|A| < p/9 + o(p)$. $(iii)$ If $|A| \gg \frac{\log\log{p}}{\sqrt{\log{p}}}p$, then $|A + A^*| \geqslant (1 - o(1))\min(2\sqrt{|A|p}, p)$. Here the constants $1/8$, $1/9$, and $2$ are the best possible. The proof involves \emph{wrappers}, subsets of a finite abelian group $G$, with which we `wrap' popular values in convolutions $A * B$ for dense sets $A, B \subseteq G$. These objects carry some special structural features, making them capable of addressing both additive-combinatorial and enumerative problems.

math.NT

On Differences of Multiplicative Functions and Solutions of the Equation $n-φ(n) = c$

We will study the solutions to the equation $f(n) - g(n) = c$, where $f$ and $g$ are multiplicative functions and $c$ is a constant. More precisely, we prove that the number of solutions does not exceed $c^{1-ε}$ when $f, g$ and solutions $n$ satisfy some certain constraints, such as $f(n) > g(n)$ for $n > 1$. In particular, we will prove the following estimate: the number of solutions to the equation $n - φ(n) = c$ is: $$ G(c + 1) + O(c^{0.75 + o(1)}), $$ where $G(k)$ is the number of ways to represent $k$ as a sum of two primes. This result is based on some properties of configurations of points and lines.

math.NT

Number of $A+B\ne C$ solutions in abelian groups and application to counting independent sets in hypergraphs

The paper deals with a problem of Additive Combinatorics. Let ${\mathbf G}$ be a finite abelian group of order $N$. We prove that the number of subset triples $A,B,C\subset {\mathbf G}$ such that for any $x\in A$, $y\in B$ and $z\in C$ one has $x+y\ne z$ equals $$ 3\cdot 4^N+N3^{N+1} + O((3-c_*)^N) $$ for some absolute constant $c_*>0$. This provides a tight estimate for the number of independent sets in a special 3-uniform linear hypergraph and gives a support for the natural conjecture concerning the maximal possible number of independent sets in such hypergraphs on $n$ vertices.

math.NT

Maximal subsets free of arithmetic progressions in arbitrary sets

We consider the problem of determining the maximum cardinality of a subset containing no arithmetic progressions of length $k$ in a given set of size $n$. It is proved that it is sufficient, in a certain sense, to consider the interval $[1,\dots, n]$. The study continues the work of Komlós, Sulyok, and Szemerédi.

math.CO