
For $0<\alpha<1$ let $V(\alpha)$ denote the supremum of the numbers $v$ such
that every $\alpha$H\"older continuous function is of bounded variation on a
set of Hausdorff dimension $v$. Kahane and Katznelson (2009) proved the
estimate $1/2 \leq V(\alpha)\leq 1/(2\alpha)$ and asked whether the upper
bound is sharp. We show that in fact $V(\alpha)=\max\{1/2,\alpha\}$. Let
$\dim_{H}$ and $\overline{\dim}_{M}$ denote the Hausdorff and upper Minkowski
dimension, respectively. The upper bound on $V(\alpha)$ is a consequence of the
following theorem. Let $\{B(t): t\in [0,1]\}$ be a fractional Brownian motion
of Hurst index $\alpha$. Then, almost surely, there exists no set $A\subset
[0,1]$ such that $\overline{\dim}_{M} A>\max\{1\alpha,\alpha\}$ and $B\colon
A\to \mathbb{R}$ is of bounded variation. Furthermore, almost surely, there
exists no set $A\subset [0,1]$ such that $\overline{\dim}_{M} A>1\alpha$ and
$B\colon A\to \mathbb{R}$ is $\beta$H\"older continuous for some
$\beta>\alpha$. The zero set and the set of record times of $B$ witness that
the above theorems give the optimal dimensions. We also prove similar
restriction theorems for deterministic selfaffine functions and generic
$\alpha$H\"older continuous functions.
Finally, let $\{\mathbf{B}(t): t\in [0,1]\}$ be a twodimensional Brownian
motion. We prove that, almost surely, there is a compact set $D\subset [0,1]$
such that $\dim_{H} D\geq 1/3$ and $\mathbf{B}\colon D\to \mathbb{R}^2$ is
nondecreasing in each coordinate. It remains open whether $1/3$ is best
possible.

The theory of dense graph limits comes with a natural sampling process which
yields an inhomogeneous variant G(n,W) of the ErdosRenyi random graph. Here we
study the clique number of these random graphs. We establish the concentration
of the clique number of G(n,W) for each fixed n, and give examples of graphons
for which G(n,W) exhibits wild longterm behavior. Our main result is an
asymptotic formula which gives the almost sure clique number of these random
graphs. We obtain a similar result for the bipartite version of the problem. We
also make an observation that might be of independent interest: Every graphon
avoiding a fixed graph is countablypartite.

Laczkovich proved that if bounded subsets $A$ and $B$ of $R^k$ have the same
nonzero Lebesgue measure and the box dimension of the boundary of each set is
less than $k$, then there is a partition of $A$ into finitely many parts that
can be translated to form a partition of $B$. Here we show that it can be
additionally required that each part is both Baire and Lebesgue measurable. As
special cases, this gives measurable and translationonly versions of Tarski's
circle squaring and Hilbert's third problem.

We prove a Borel version of the local lemma, i.e. we show that, under
suitable assumptions, if the set of variables in the local lemma has a
structure of a Borel space, then there exists a satisfying assignment which is
a Borel function. The main tool which we develop for the proof, which is of
independent interest, is a parallel version of the MoserTardos algorithm which
uses the same random bits to resample clauses that are far enough in the
dependency graph.

Given an action of a group $\Gamma$ on a measure space $\Omega$, we provide a
sufficient criterion under which two sets $A, B\subseteq \Omega$ are measurably
equidecomposable, i.e., $A$ can be partitioned into finitely many measurable
pieces which can be rearranged using the elements of $\Gamma$ to form a
partition of $B$. In particular, we prove that every bounded measurable subset
of $R^n$, $n\ge 3$, with nonempty interior is measurably equidecomposable to a
ball via isometries. The analogous result also holds for some other spaces,
such as the sphere or the hyperbolic space of dimension $n\ge 2$.

It is known that if the underlying iterated function system satisfies the
open set condition, then the upper box dimension of an inhomogeneous
selfsimilar set is the maximum of the upper box dimensions of the homogeneous
counterpart and the condensation set. First, we prove that this `expected
formula' does not hold in general if there are overlaps in the construction. We
demonstrate this via two different types of counterexample: the first is a
family of overlapping inhomogeneous selfsimilar sets based upon Bernoulli
convolutions; and the second applies in higher dimensions and makes use of a
spectral gap property that holds for certain subgroups of $SO(d)$ for $d\geq
3$.
We also obtain new upper bounds for the upper box dimension of an
inhomogeneous selfsimilar set which hold in general. Moreover, our
counterexamples demonstrate that these bounds are optimal. In the final section
we show that if the \emph{weak separation property} is satisfied, ie. the
overlaps are controllable, then the `expected formula' does hold.

Following Davies, Elekes and Keleti, we study measured sets, i.e. Borel sets
$B$ in $\mathbb{R}$ (or in a Polish group) for which there is a translation
invariant Borel measure assigning positive and \sigmafinite measure to $B$. We
investigate which sets can be written as a (disjoint) union of measured sets.
We show that every Borel nullset $B\subset \mathbb{R}$ of the second category
is larger than any nullset $A\subset \mathbb{R}$ in the sense that there are
partitions $B=B_1\cup B_2$, $A=A_1\cup A_2$ and gauge functions $g_1, g_2$ such
that the Hausdorff measures satisfy $H^{g_i}(B_i)=1$ and $H^{g_i}(A_i)=0$
($i=1,2$). This implies that every Borel set of the second category is a union
of two measured sets.
We also present Borel and compact sets in $\mathbb{R}$ which are not a union
of countably many measured sets. This is done in two steps. First we show that
nonlocally compact Polish groups are not a union of countably many measured
sets. Then, to certain Banach spaces we associate a Borel and/or \sigmacompact
additive subgroup of $\mathbb{R}$ which is not a union of countably many
measured sets.
It is also shown that there are measured sets which are null or
non\sigmafinite for every Hausdorff measure of arbitrary gauge function.

We give a sketch of proof that any two (Lebesgue) measurable subsets of the
unit sphere in $R^n$, for $n\ge 3$, with nonempty interiors and of the same
measure are equidecomposable using pieces that are measurable.

A famous conjecture of Lov\'asz states that every connected vertextransitive
graph contains a Hamilton path. In this article we confirm the conjecture in
the case that the graph is dense and sufficiently large. In fact, we show that
such graphs contain a Hamilton cycle and moreover we provide a polynomial time
algorithm for finding such a cycle.

S.Janson [Poset limits and exchangeable random posets, Combinatorica 31
(2011), 529563] defined limits of finite posets in parallel to the emerging
theory of limits of dense graphs.
We prove that each poset limit can be represented as a kernel on the unit
interval with the standard order, thus answering an open question of Janson. We
provide two proofs: realanalytic and combinatorial. The combinatorial proof is
based on a Szemereditype Regularity Lemma for posets which may be of
independent interest.
Also, as a byproduct of the analytic proof, we show that every atomless
ordered probability space admits a measurepreserving and almost
orderpreserving map to the unit interval.

More than 80 years ago Kolmogorov asked the following question. Let
$E\subseteq \mathbb{R}^{2}$ be a measurable set with $\lambda^{2}(E)<\infty$,
where $\lambda^2$ denotes the twodimensional Lebesgue measure. Does there
exist for every $\varepsilon>0$ a contraction $f\colon E\to \mathbb{R}^{2}$
such that $\lambda^{2}(f(E))\geq \lambda^{2}(E)\varepsilon$ and $f(E)$ is a
polygon? We answer this question in the negative by constructing a bounded,
simply connected open counterexample. Our construction can easily be modified
to yield the analogous result in higher dimensions.

Let $X$ be a Polish space. We prove that the generic compact set $K\subseteq
X$ (in the sense of Baire category) is either finite or there is a continuous
gauge function $h$ such that $0<\mathcal{H}^{h}(K)<\infty$, where
$\mathcal{H}^h$ denotes the $h$Hausdorff measure. This answers a question of
C. Cabrelli, U. B. Darji, and U. M. Molter. Moreover, for every weak
contraction $f\colon K\to X$ we have $\mathcal{H}^{h} (K\cap f(K))=0$. This is
a measure theoretic analogue of a result of M. Elekes.

We prove that a compact metric space (or more generally an analytic subset of
a complete separable metric space) of Hausdorff dimension bigger than $k$ can
be always mapped onto a $k$dimensional cube by a Lipschitz map. We also show
that this does not hold for arbitrary separable metric spaces.
As an application we essentially answer a question of Urba\'nski by showing
that the transfinite Hausdorff dimension (introduced by him) of an analytic
subset $A$ of a complete separable metric space is the integer part of $\dim_H
A$ if $\dim_H A$ is finite but not an integer, $\dim_H A$ or $\dim_H A1$ if
$\dim_H A$ is an integer and at least $\omega_0$ if $\dim_H A=\infty$.

We study the following two problems:
(1) Given $n\ge 2$ and $\al$, how large Hausdorff dimension can a compact set
$A\su\Rn$ have if $A$ does not contain three points that form an angle $\al$?
(2) Given $\al$ and $\de$, how large Hausdorff dimension can a %compact
subset $A$ of a Euclidean space have if $A$ does not contain three points that
form an angle in the $\de$neighborhood of $\al$?
An interesting phenomenon is that different angles show different behaviour
in the above problems. Apart from the clearly special extreme angles 0 and
$180^\circ$, the angles $60^\circ,90^\circ$ and $120^\circ$ also play special
role in problem (2): the maximal dimension is smaller for these special angles
than for the other angles. In problem (1) the angle $90^\circ$ seems to behave
differently from other angles.

Let us say that an element of a given family $\A$ of subsets of $\R^d$ can be
reconstructed using $n$ test sets if there exist $T_1,...,T_n \subset \R^d$
such that whenever $A,B\in \A$ and the Lebesgue measures of $A \cap T_i$ and $B
\cap T_i$ agree for each $i=1,...,n$ then $A=B$. Our goal will be to find the
least such $n$.
We prove that if $\A$ consists of the translates of a fixed reasonably nice
subset of $\R^d$ then this minimum is $n=d$. In order to obtain this result we
reconstruct a translate of a fixed function using $d$ test sets as well, and
also prove that under rather mild conditions the measure function $f_{K,\theta}
(r) = \la^{d1} (K \cap \{x \in \RR^d : <x,\theta> = r\})$ of the sections of
$K$ is absolutely continuous for almost every direction $\theta$. These proofs
are based on techniques of harmonic analysis.
We also show that if $\A$ consists of the magnified copies $rE+t$ $(r\ge 1,
t\in\R^d)$ of a fixed reasonably nice set $E\subset \R^d$, where $d\ge 2$, then
$d+1$ test sets reconstruct an element of $\A$. This fails in $\R$: we prove
that an interval, and even an interval of length at least 1 cannot be
reconstructed using 2 test sets.
Finally, using randomly constructed test sets, we prove that an element of a
reasonably nice $k$dimensional family of geometric objects can be
reconstructed using $2k+1$ test sets. A example from algebraic topology shows
that $2k+1$ is sharp in general.

The main result of this paper is the following. Given countably many
multivariate polynomials with rational coefficients and maximum degree $d$, we
construct a compact set $E\subset \R^n$ of Hausdorff dimension $n/d$ which does
not contain finite point configurations corresponding to the zero sets of the
given polynomials.
Given a set $E\subset \R^n$, we study the angles determined by three points
of $E$. The main result implies the existence of a compact set in $\R^n$ of
Hausdorff dimension $n/2$ which does not contain the angle $\pi/2$. (This is
known to be sharp if $n$ is even.) We show that there is a compact set of
Hausdorff dimension $n/8$ which does not contain an angle in any given
countable set. We also construct a compact set $E\subset \R^n$ of Hausdorff
dimension $n/6$ for which the set of angles determined by $E$ is Lebesgue null.
In the other direction, we present a result that every set of sufficiently
large dimension contains an angle $\epsilon$ close to any given angle.

A \emph{hull} of $A \subset [0,1]$ is a set $H$ containing $A$ such that
$\lambda^*(H)=\lambda^*(A)$. We investigate all four versions of the following
problem. Does there exist a monotone (wrt. inclusion) map that assigns a
Borel/$G_\delta$ hull to every negligible/measurable subset of $[0,1]$?
Three versions turn out to be independent of ZFC (the usual ZermeloFraenkel
axioms with the Axiom of Choice), while in the fourth case we only prove that
the nonexistence of a monotone $G_\delta$ hull operation for all measurable
sets is consistent. It remains open whether existence here is also consistent.
We also answer a question of Z. Gyenes and D. P\'alv\"olgyi which asks if
monotone hulls can be defined for every chain (wrt. inclusion) of measurable
sets. We also comment on the problem of hulls of all subsets of $[0,1]$.

Let K be a selfsimilar or selfaffine set in R^d, let \mu be a selfsimilar
or selfaffine measure on it, and let G be the group of affine maps,
similitudes, isometries or translations of R^d. Under various assumptions (such
as separation conditions or we assume that the transformations are small
perturbations or that K is a so called Sierpinski sponge) we prove theorems of
the following types, which are closely related to each other;
Nonstability: There exists a constant c<1 such that for every g\in G we have
either \mu(K\cap g(K)) <c \mu(K) or K\subset g(K).
Measure and topology: For every g\in G we have \mu(K\cap g(K)) > 0 \iff int_K
(K\cap g(K)) is nonempty (where int_K is interior relative to K).
Extension: The measure \mu has a Ginvariant extension to R^d.
Moreover, in many situations we characterize those g's for which \mu(K\cap
g(K) > 0, and we also get results about those $g$'s for which $g(K)\su K$ or
$g(K)\supset K$ holds.