SearcharxivSearch

arXiv subjects

Tomasz Kowalczyk

Publications and source records attributed to Tomasz Kowalczyk.

8 recordsLinked to original sources

A note on the real Jacobian conjecture in degree 7

Let $(p,q)$ be a Jacobian pair. We show that the real Jacobian conjecture holds if the degree of $p$ is 7 and the highest degree homogenous part is of the form $αx^7 + βx^6y$ for $α^2+β^2 \neq 0$. We then show that there are no atypical Jacobian pairs such that $\mathrm{deg} \, p =7$ and $\mathrm{deg}\, q$ is even and coprime with 7.

math.AG

Sums of squares on curves and surfaces

We study sums of higher even powers in the coordinate rings of singular planar curves $x^M=y^m$ for coprime positive integers $m 1$, can be of codimension $2$, contrary to the quadratic case.

math.AC

Sums of squares of regular functions on rational surfaces

We study the sums of squares on cylinders of the form $X \times \mathbb{A}_K$ for a (weakly) factorial curve $C$. We prove the equality of the Pythagoras numbers of the ring of regular functions on the cylinder with that of the field of rational functions. We then apply these results to the case of (uniformly) rational varieties. We show that if $X$ is a nonsingular rational algebraic surface over the reals, then the Pythagoras number of the ring of regular functions on $X$ is bounded above by 12.

math.AG

On Waring numbers of henselian rings

Let $n>1$ be a positive integer. Let $R$ be a henselian local ring with residue field $k$ of $n$th level $s_n(k)$. We give some upper and lower bounds for the $n$th Waring number $w_n(R)$ in terms of $w_n(k)$ and $s_n(k)$. In large number of cases we are able to compute $w_n(R)$. Similar results for the $n$th Waring number of the total ring of fractions of $R$ are obtained. We then provide applications. In particular we compute $w_n(\mathbb{Z}_p)$ and $w_n(\mathbb{Q}_p)$ for $n\in\{3,4,5\}$ and any prime $p$.

math.AC

On higher Pythagoras numbers of polynomial rings

We show that the higher Pythagoras numbers for the polynomial ring are infinite $p_{2s}(K[x_1,x_2,\dots,x_n])=\infty$ provided that $K$ is a formally real field, $n\geq2$ and $s\geq 1$. This almost fully solves an old question \cite[Problem 8]{cldr1982}. The remaining open cases are precisely $n=1$ and $s>1$. Moreover, we study in detail the cone of binary octics that are sums of fourth powers of quadratic forms. We determine its facial structure as well as its algebraic boundary. This can also be seen as sums of fourth powers of linear forms on the second Veronese of $\mathbb{P}^1$. As a result, we disprove a conjecture of Reznick \cite[Conjecture 7.1]{reznick2011} which states that if a binary form $f$ is a sum of fourth powers, then it can be written as $f=f_1^2+f_2^2$ for some nonnegative forms $f_1,f_2$.

math.AG

Sums of squares on hypersurfaces

We show that the Pythagoras number of rings of type $\mathbb{R}[x,y, \sqrt{f(x,y)}]$ is infinite, provided that the polynomial $f(x,y)$ satisfies some mild conditions.

math.AG

Sums of even powers of k-regulous functions

We provide an example of a nonnegative $k$-regulous function on $\mathbb{R}^n$ for $k\geq 1$ and $n \geq 2$ which cannot be written as a sum of squares of $k$-regulous functions. We then obtain lower bounds for Pythagoras numbers $p_{2d}(\mathcal{R}^k(\mathbb{R}^n))$ of $k$-regulous functions on $\mathbb{R}^n$ for $k\geq 1$ and $n\geq 2$. We also prove that the second Pythagoras number of the ring of $0$-regulous functions $\mathcal{R}^0(X)$ on an irreducible $0$-regulous affine variety $X$ is finite and bounded from above by $2^{\dim X}$.

math.AG

Blown-up Čech cohomology and Cartan's Theorem B on real algebraic varieties

We introduce a concept of blown-up Čech cohomology for coherent sheaves of homological dimension $\leq 1$ and some quasi-coherent sheaves on a non-singular real affine variety. Its construction involves a directed set of multi-blowups. We establish, in particular, long exact cohomology sequence and Cartan's Theorem B. Finally, some applications are provided, including universal solution to the first Cousin problem (after blowing up).

math.AG