## Geometric function and mapping theory

### dr hab. Tomasz Adamowicz

Thursday 14.15-15.45, room 106Academic year 2023/2024

FALL SEMESTER

**5-Oct: Organizational meeting**

**12-Oct: No seminar, conference MEGA (Będlewo)**

**19-Oct**: **No seminar, the Stean Banach Medal ceremony**

**26-Oct: Michał Łasica (IM PAN)**

**Title: On jump discontinuities of minimizers in variational denoising**

**Abstract: ** We consider the minimization problem for a class of convex integral functionals composed of two terms:

-- a regularizing term of linear growth in the gradient,

-- and a fidelity term penalizing the distance from a noisy datum.

Such functionals attain their minima on the BV space of functions of bounded variation, in particular the minimizers may exhibit jump discontinuities. This is a desirable feature in image denoising, corresponding to sharp contours. We show an estimate on the location and size of jumps of the minimizers in terms of data. Our method works for a large class of regularizers under a mild assumption of differentiability along inner variations, and applies in the vectorial setting, corresponding to color images.

This is joint work with Antonin Chambolle.

**2-Nov: No seminar,**** All Souls' Day**

**9-Nov: Ben Warhurst** **(MIM UW),**

**Title: Moduli for sublaplacians on the second Heisenberg group**

**Abstract: ** A general sublaplacian is an operator of the form div_H(M grad_H f) where div_H is a horizontal divergence, M is a symmetric positive definite matrix acting on the horizontal bundle, and grad_H is a horizontal gradient. In the Euclidean setting one can always find a change of coordinates that brings such an operator into the standard form div(grad f) using the symmetric square root C of M, however this is not always possible on a stratified group since C must also extend to an automorphism of the Lie algebra of the group. If the group is free then extending C to an automorphism is not a problem and the symmetric square root works. The second Heisenberg group is perhaps the simplest nonfree stratified group. In this case we employ a recently developed theory of horizontal jets to reveal that the classes of contact equivalent sublaplacians are uniquely determined by a positive real parameter.

**16-Nov: Jan Rozendaal (IM PAN) **

**Title: HARDY SPACES FOR FOURIER INTEGRAL OPERATORS**

**Abstract:** It is well known that the wave operators $\cos(t\sqrt{-\Delta})$ and $\sin(t\sqrt{-\Delta})$ are not bounded on $L^{p}(\R^{n})$, for $n\geq 2$ and $1\leq p\leq \infty$, unless $p=2$ or $t=0$. In fact, for $1

In this talk, I will introduce a class of Hardy spaces $\mathcal{H}^{p}_{FIO}(\R^{n})$, for $1\leq p\leq\infty$, on which suitable Fourier integral operators are bounded. These spaces also satisfy Sobolev embeddings that allow one to recover the optimal boundedness results for Fourier integral operators on $L^{p}(\Rn)$.

In fact, the invariance of these spaces under Fourier integral operators allows for iterative constructions that are not possible when working directly on $L^{p}(\Rn)$. I will also mention the connection of these spaces to the local smoothing conjecture.

The research leading to these results has received funding from the Norwegian Financial Mechanism 2014-2021, grant 2020/37/K/ST1/02765.

**21-Nov, r. 106: Grzegorz Łysik (UJK) **

**Title: **Smoothness of the Dunkl analytic functions

**Abstract:** For the reflection group $W$ associated with a finite root system and a $W$-invariant weight function $\omega_\kappa$ Dunkl introduced a differential-difference operators $T_j$, $j = 1, . . . .n$, and the Dunkl Laplacian $\Delta_\kappa=\sum_{j=1}^nT_j ^2$.

A continuous function on a $W$-invariant set $\Omega$ is called Dunkl analytic if its mean value function over balls in $\Omega$ of radius $R$ with respect to the measure $\omega_\kappa(x)dx$ is convergent for small $R > 0$. During the talk we shall show that Dunkl analytic functions are smooth and real analytic in the case $W=\mathbb{Z}_2^n$.

**30-Nov: Marcin Gryszówka (IM PAN) **

**Title: **Carleson measure and inequality between square and nontangential maximal functions in Heisenberg group

**Abstract: **It is a well established result that the square function and nontangential maximal function are comparable in the Euclidean space for a good enough domain Ω. It is a crucial part of a proof that |∇u|^{2}dist(x, ∂Ω) is a Carleson measure for a bounded harmonic function u : Ω → R. We want to investigate inequality between the square function and nontangential maximal function in the setting of Heisenberg group and obtain a proper Carleson measure. In the talk I want to recall the proof in the Euclidean case and see what is necessary to get the desired result in Heisenberg group and why we can get a Carleson measure there.

**7-Dec: Maria J. Gonzalez (University of Cádiz, Spain)**

**Title:** HARDY SPACES AND QUASIREGULAR MAPPINGS

**Abstract: **We study Hardy spaces for quasiregular mappings on the unit ball which satisfy appropriate growth and multiplicity conditions. Under these conditions we recover several classical results for analytic functions and quasiconformal mappings. In particular, we characterize Hardy spaces in terms of non-tangential limit functions and non-tangential maximal functions of

quasiregular mappings. Among applications we show that every quasiregular map in our class belongs to some Hardy space.

A key difference between the previously known results for quasiconformal mappings and our setting is the role of multiplicity conditions and the growth of mappings that need not be injective. (Joint work with T. Adamowicz)

**14-Dec: Marcin Walicki (MiNI PW)**

**Title:** Harmonic and p-harmonic maps between Riemannian manifolds.

**Abstract: **During the talk, we recall the definition of the energy of maps between Riemannian manifolds as well as what it means for a map to be harmonic or, more generally, p-harmonic. Furthermore, we will provide some geometric interpretations of the energy of a map and lastly, we will discuss some theorems regarding the existence, uniqueness, regularity, and the Dirichlet problem in the case of the map being p-harmonic.

**21-Dec, 28-Dec, 4-Jan: No seminar, Winter holidays**

**11-Jan: Remy Rodiac (MIM UW), joint seminar with DE Seminar**

**Title: **Large systems of interacting points at equilibrium in the plane

**Abstract:** In this talk I will consider equilibrium solutions of a system of ODEs describing N points interacting via a Coulomb potential in the plane. When N is large we expect the limit of the empirical measure of the positions of the points to satisfy a system of PDEs. I will describe the difficulties to understand the limit N goes to infinity (transform the equilibrium relations into a non-linear PDE and pass to the limit in this PDE) and explain how we can solve them to obtain the result.

SPRING SEMESTER

**21-Mar: ****Marcin Gryszówka (IM PAN)**

**Title:** Singular integral operators - Introduction

**Abstract**: I will present a review of results concerning singular integral operators. In particular I will show a theorem by Stein stating when an operator given by a convolution is bounded in Lp. What is more, I will talk about some results connecting boundedness of certain integral operators and uniform rectifiability of a domain. I will consider both a Euclidean case and a case of the Heisenberg group.

**4-Apr: ****Marcin Gryszówka (IM PAN)**

**Title:** Singular integral operators - Introduction, pt. 2

**Abstract**: I will present a review of results concerning singular integral operators. In particular I will show a theorem by Stein stating when an operator given by a convolution is bounded in Lp. What is more, I will talk about some results connecting boundedness of certain integral operators and uniform rectifiability of a domain. I will consider both a Euclidean case and a case of the Heisenberg group.

**11-Apr: Abhishek Ghosh (IM PAN)**

**Title: **Maximal functions: unweighted and weighted estimates*.*

**Abstract**: Maximal functions are of great interest in Harmonic Analysis, PDEs, Ergodic theory and other areas of Mathematics and their study goes back to the time of Hardy and Littlewood. In this talk, firstly, we shall see how they arise very naturally and how to obtain useful estimates for maximal functions on euclidean spaces using covering lemmas. The rest of the talk is dedicated to the analysis of two typical scenarios: the first one is the study of maximal functions in the absence of covering arguments, and the second is the study of maximal functions arising from wave equations.

**9-May: First meeting of the Seminar in Geometric Analysis**

__Schedule of the seminar:__

**16-May: Łukasz Chomienia (MIM UW)**

**Title: **Geometric approach to regularity of solutions in the low-dimensional framework

**Abstract**: The talk pertains to elliptic-type partial differential equations examined in the context of "thin" subsets of Euclidean space, specifically, the so-called low-dimensional structures. Various regularity aspects of weak solutions are the primary focus. By utilizing particular geometric extensions of functions supported on "thin" structures, we attain higher Sobolev regularity on every component manifold of the low-dimensional structure. However, this regularity type is too weak to fully capture a higher-order interplay between various components of the structure.

We upgrade the regularity type to obtain membership of weak solutions in the domain of the low-dimensional second-order operator. To accomplish this, we demonstrate that a particular normal to the low-dimensional structure vector field exists that regularizes the structure-related gradient. Special attention will be devoted to geometric constructions related both to the first-and second-order lower dimensional setting.

.

Academic year 2022/2023

SPRING SEMESTER

**16-Mar: Marcin Gryszówka (IM PAN)**

**Title: Fatou type theorem for harmonic functions in Heisenberg groups**

**Abstract: ** In the Euclidean setting the well known Fatou theorem states that a nontangential limit of a positive harmonic function exists at a.e. point of a boundary. It turns out that if one changes a Euclidean metric to Carnot- Caratheodory metric the same theorem holds for nontangentially accessible domains (NTA domains). Moreover, it is true for more general operators than only the harmonic one. I will discuss the aforementioned results and present techniques used to prove them.

**23-Mar: Marcin Walicki (MiNI PW)**

The seminar moved to 20-Apr and 27-Apr.

**30-Mar: Alberto Maione (Albert-Ludwigs-University of Freiburg, Germany)**

**Title: VARIATIONAL CONVERGENCES FOR FUNCTIONALS AND DIFFERENTIAL OPERATORS DEPENDING ON VECTOR FIELDS**

**Abstract: **In this talk I present results concerning variational convergences for functionals and differential operators depending on a family of Lipschitz continuous vector fields X, that may not satisfy the well-known Hormander condition. Classic examples of such families are the Euclidean gradient, the Grushin vector fields and the Heisenberg vector fields (as prototype of any Carnot group). This setting was introduced by Folland and Stein and found numerous applications in the literature. The convergences taken into account date back to the 70's and are Gamma-convergence, introduced by De Giorgi and Franzoni, dealing with functions and functionals, and G-convergence, or H-convergence, whose theory was initiated by De Giorgi and Spagnolo and developed by Murat and Tartar. This last convergence deals with differential operators. The main result presented today is a Gamma-compactness theorem, which ensures that sequences of integral functionals depending on X, with standard regularity and growth conditions, Gamma-converge (up to subsequences) in the strong topology of Lp (p > 1) to a functional belonging to the same class, that is, satisfying the same regularity and growth conditions and representable in an integral form.

As an interesting consequence of the previous result, I finally show that the class of linear differential operators in X-divergence form is closed in the topology of the H-convergence. The variational technique adopted to this aim relies on a new approach, introduced by N. Ansini, G. Dal Maso and C. I. Zeppieri.

This research is done in collaboration with Andrea Pinamonti and Francesco Serra Cassano (University of Trento), Fabio Paronetto (University of Padova) and Eugenio Vecchi (University of Bologna).

**6-Apr: **No seminar (Good Thursday).

**13-Apr: **No seminar (Fribourg conference MAPS).

**20-Apr: Marcin Walicki (MiNI PW), pt. 1.**

**Title: **"Geometrical objects on the Grushin plane".

**Abstract: **During the talk, we will discuss various intrinsic geometric objects on the Grushin plane with special emphasis on intrinsic curvature, its representations and geometry of the level sets with zero intrinsic curvature. The presentation is based on a paper by Ferrari and Valdinoci ,,Geometric PDEs in the Grushin Plane: Weighted Inequalities and Flatness of Level Sets''.

**27-Apr:** **Marcin Walicki (MiNI PW), pt. 2.**

**Title:**,,The Sobolev-Poincare inequality for solutions of Allen-Cahn equation in the Grushin plane''

**Abstract: **We continue our exploration of geometrical objects on the Grushin plane with some of their applications. It turns out that stable solutions of the Allen-Cahn equation in the Grushin plane satisfy a Sobolev-Poincare inequality. Such inequality bounds the weighted L^2-norm of a test function by a weighted L^2-norm of its gradient, and the weights involve geometry-related quantities.

**4-May: No seminar due to the Long May Weekend.**

**11-May: Jan Rozendaal (IM PAN), seminar postponed**

**Title:** HARDY SPACES FOR FOURIER INTEGRAL OPERATORS

**Abstract:** It is well known that the wave operators $\cos(t\sqrt{-\Delta})$ and $\sin(t\sqrt{-\Delta})$ are not bounded on $L^{p}(\R^{n})$, for $n\geq 2$ and $1\leq p\leq \infty$, unless $p=2$ or $t=0$. In fact, for $1 these operators are bounded from $W^{2s(p),p}(\R^{n})$ to $L^{p}(\R^{n})$ for $s(p):=\frac{n-1}{2}|\frac{1}{p}-\frac{1}{2}|$, and this exponent cannot be improved. This phenomenon is symptomatic of the behavior of Fourier integral operators, a class of oscillatory operators which includes wave propagators.

In this talk, I will introduce a class of Hardy spaces $\mathcal{H}^{p}_{FIO}(\R^{n})$, for $p\in[1,\infty]$, on which Fourier integral operators of order zero are bounded. These spaces also satisfy Sobolev embeddings that allow one to recover the optimal boundedness results for Fourier integral operators on $L^{p}(\Rn)$.

However, beyond merely recovering existing results, the invariance of these spaces under Fourier integral operators allows for iterative constructions that are not possible when working directly on $L^{p}(\Rn)$. For example, they can be used to obtain the optimal fixed-time $L^{p}$ regularity for wave equations with rough coefficients. Moreover, these spaces are connected to the phenomenon of local smoothing, which in turn is relevant for both harmonic and geometric analysis.

This talk is based on joint work with Andrew Hassell and Pierre Portal (Australian National University), and Zhijie Fan, Naijia Liu, Liang Song and Lixin Yan (Sun Yat-Sen University).

**18-May: Wojciech Kryński (IM PAN).**

**Title: ** Variational approach to conformal geodesics

**Abstract: **It is known that geodesics are distance-minimizing curves. A conformal structure on a manifold is a class of metrics given up to a factor. Consequently, the notion of geodesics becomes meaningless on conformal manifolds. However, there is a conformally invariant equation of the third order whose solutions constitute a well-defined class of curves. In the flat case, these are circles. In general, they are called conformal geodesics. Since the equation is of the third order, it is unclear how to interpret it in terms of a variational principle (the Euler-Lagrange equations are of even order). In this talk I will show how to overcome these difficulties. I will also discuss a variational formulation of the Schwarzian equation.

**25-May: Zofia Grochulska (MIM UW).**

**Title: **Approximately differentiable homeomorphisms with prescribed derivative: a happy ending to a long construction.

**Abstract: **One of the main reasons to be interested in a.e. approximately differentiable mappings is the fact proved by Federer that they are admissible for area formula. I will talk about properties of such maps and focus on the question of prescribing the derivative of approximately differentiable homeomorphisms and how it is connected with Oxtoby-Ulam theorem about homeomorphic measures. I will show a stronger version of the theorem presented in the fall: that under some mild assumptions on a map T defined on the unit cube Q with values in GL(n), we can find an approximately differentiable homeomorphism, which equals identity on the boundary of Q and whose approximate derivative equals T a.e. This is joint work with Paweł Goldstein and Piotr Hajłasz.

**1-June: Colloquium by Alexandre Eremenko, **14:30-15:20, MIMUW, room 5440

see https://www.impan.pl/konferencje/bcc/2023/23-simons-v/eremenko-lecture.pdf

**8-June: **No seminar (Corpus Christi).

**15-June: Michał Miśkiewicz (MIM UW).**

**22-June: Marco Capolli (IM PAN).**

FALL SEMESTER

**15-Sep: A. Pinamonti (Trento, Italy)**

**Title: Whitney Extension and Lusin Approximation for Horizontal Curves in the Heisenberg Group**

**Abstract: ** Whitney extension results characterize when one can extend a mapping from a compact subset to a smooth mapping on a larger space. Lusin approximation results give conditions under which one can approximate a rough map by a smoother map after discarding a set of small measure. We first recall relevant results in the Euclidean setting, then describe recent work extending them to horizontal curves in the Heisenberg group. We focus on C^m curves.

**6-Oct: Organizational meeting**

**13-Oct: Ben Warhurst (University of Warsaw)**

**Title:**** Horizontal jet spaces and equivalence**

**Abstract: **This talk will consider the idea of Horizontal jets for functions on stratified groups and their application to questions of equivalence.

**20-Oct: Elefterios Soultanis (Radboud University, Holland)**

**Title: The tensorization problem for Sobolev spaces**

**Abstract: **If a function, defined on the product of two spaces X and Y, has p-integrable partial derivatives in the X and Y -directions, is it a Sobolev function? When X and Y are Euclidean this is an easy exercise, but the question becomes very difficult in the absence of smooth structures, and is known as the tensorization problem for Sobolev spaces. In this talk I focus on metric spaces X and Y, where it is an open problem. I'll discuss some known cases and recent approaches to the tensorization problem for metric measure spaces.

**27-Oct: Marco Capolli (IM PAN)**

**Title: Directional Derivatives and Geodesics in the Laakso Space**

**Abstract: **A Universal Differentiability Set, UDS for short, is a set such that every Lipschitz function is differentiable in at least one point of the set. The existence of measure zero UDS, which is strictly related to Rademacher's theorem, has already been established in the Euclidean setting and for certain classes of Carnot groups. The key technique underlying the construction of UDS is a result that relates maximality of a directional derivative for a Lipschitz function with differentiability. In this seminar we will investigate this connection and we will try to establish how things change when we work in a nonlinear setting. To do so we will focus our attention on the Laakso space, one of the best known and most interesting instances of a metric measure space which admits a differentiable structure but not a linear structure. An introduction to Laakso spaces and their structure will be provided during the seminar.

**3-Nov: Remy Rodiac (Université Paris-Saclay)**

**Title: Towards a relaxed energy for the neo-Hookean energy in the axially symmetric setting**

**Abstract:**** **We consider the problem of minimizing hyperelastic energies in 3D. It is known that if the energy space allows for maps with cavitation (creation of holes), without penalization of the created surface, the hyperelastic energies are not lower semi-continuous for the weak convergence. We can then try to find minimizers in a subset of maps not allowing for cavitation, but for the neo-Hookean energy this condition is not closed under the weak convergence in H^1. I will describe some properties of weak limits of minimizng sequences and give a lower-bound on the relaxed energy for the neo-Hookean problem. This is a joint work (in progress) with Marco Barchiesi, Duvan Henao and Carlos Mora-Corral.

**10-Nov: Marcin Gryszówka (IM PAN)**

**Title: Epsilon-approximation on Riemannian manifolds**

**Abstract: **It is already known that bounded harmonic functions defined on Lipschitz bounded domains in Euclidean space are epsilon-approximable. It means that for a bounded harmonic function u and a given epsilon one can find a function of bounded variation such that its distance from u in supremum norm is less than epsilon and the norm of its gradient is a Carleson measure. We will prove that in the setting of Riemannian manifold the same holds. That property is essential in proving Quantitative Fatou Property on Riemannian manifold.

**17-Nov: No seminar**

** 24-Nov: Marcin Walicki (PW)**

**Title: **Quasiconformal mappings on the Grushin plane, p.I

**Abstract:** Quasiconformal mappings are generalizations of conformal mappings. There are 3 main definitions of quasiconformality: metric, geometric and analytic one. Their equivalence is known in various metric measure space settings. However, until recent work by Ackermann (2010) and Gartland, Jung, and Romney (2016), modern theory was unable to cover the case of the Grushin plane. The talk will start with basic definitions of quasiconformality in Euclidean spaces and example of the appearance of quasiregular mapping in Physics. We will proceed with defining quasiconformal mappings in a more general sense, outline general knowledge about the equivalence of definitions in metric measure spaces, and finally, provide recent development of this theory on the Grushin plane.

**1-Dec: David Tewodrose (Nantes Université)**

**Title: ** Torus stability of closed Riemannian manifolds under Kato bounds on the Ricci curvature

**Abstract: ** In this talk, I will present recent results obtained with Gilles Carron (Nantes Université) and Ilaria Mondello (Université Paris-Est Créteil). We work in the setting of closed Riemannian manifolds satisfying a so-called Kato bound, that is to say, the negative part of the optimal lower bound on the Ricci curvature lies in some suitable Kato class. I will explain how this assumption provides good analytic and geometric control on the manifold, and how this yields that a closed Riemannian manifold satisfying a strong Kato bound with maximal first Betti number is necessarily diffeomorphic to a torus: this extends a result by Colding and Cheeger-Colding obtained in the context of a lower bound on the Ricci curvature.

**8-Dec: Maria J. Gonzalez (University of Cádiz, Spain)**

**Title: **Quasiconformal properties of Q_{p,0} curves and Dirichlet-type curves

**Abstract: **Let Γ be a closed Jordan curve, and f the conformal mapping that sends the unit disc D onto the interior domain of Γ. If log f' belongs to the Dirichlet space D, we call Γ a Weil-Petersson curve. The purpose of this talk is to discuss extensions of recent results, obtained by G. Cui and Ch. Bishop in the case of Weil-Petersson curves, to the case when log f′ belongs to either some Qp,0, space, for 0 < p ≤ 1, or to some weighted-Dirichlet space contained in D. More precisely, we will characterize the quasiconformal extensions of f , and describe some of the geometric properties of Γ, that arise in this context.

**15-Dec: Zofia Grochulska (MIM UW)**

**Title: TBA**

**Abstract:**

**22-Dec, 29-Dec, 5-Jan: No seminar, Winter holidays **

**12-Jan:**

**Title: TBA**

**Abstract:**

**19-Jan:**

**Title: TBA**

**Abstract:**

**26-Jan:**

**Title: TBA**

**Abstract:**

** **Academic year 2021/2022

SPRING SEMESTER

**7-Mar: A. Kijowski (OIST, Japan)**

**Title: **"Quasiconvex envelope in the Heisenberg group"

**Abstract: ** In my presentation I will discuss problems related to (quasi)convexity of sets and functions in the Heisenberg group. I will start with introducing different definitions of convex sets in this setting and definitions of convex and quasiconvex functions. I will demonstrate zero, first and second order characterization of these notions. Finally, I will establish an iterative construction of the quasiconvex envelope of a continuous function.

The talk is based on joint work with Qing Liu and Xiaodan Zhou.

**17-Mar: T. Cieślak (IMPAN)**

**Title: ** "Phragmen-Lindelof estimates for a problem occurring in magnetostatic levitation"

**Abstract: W**e shall consider a linear second-order elliptic operator in an infinite strip. The considered problem is taken from the model of magnetostatic levitation. Engineers wrote down a particular solution to the problem, which allowed them computing a lift and a drag. A natural question appears: is the solution unique? We answer the above question in an affirmative way. To this end we use an approach of E.Hopf dating back to 1950s. Moreover, the problem has direct connections to the Laplacian Eigenproblem in a strip in the class of differentiable bounded functions as well as a degenerate elliptic problem on a half-space.

The talk is based on a joint paper with B. Bieganowski and J. Siemianowski.

**24-Mar: K. Mazowiecka (MIM UW)**

**Title: **"Fractional harmonic maps in homotopy classes"

**Abstract: **We extend the seminal work of Sacks and Uhlenbeck into the fractional framework. We look for maps between manifolds M and N that minimize the critical Sobolev W(s,n/s) energy, where n is the dimension of the manifold M in the domain. We prove that in the case when the target manifold N has trivial n-th homotopy group then in every homotopy class of maps between an n-manifold M to N there exists a minimizing fractional harmonic map. In case when the n-th homotopy group of the target manifold is not trivial we prove that there exists a generating set for the n-th homotopy group of the target manifold N consisting of minimizing fractional maps.

We develop new tools which are interesting on their own, such as a removability result for point-singularities and a balanced energy estimate for non-scaling invariant energies.

**31-Mar: K. Mazowiecka (MIM UW)**

**Title: **"Fractional harmonic maps in homotopy classes, part II"

**Abstract: **This is the continuation of the talk given in the previous week.

**7-Apr: E. Soultanis (Radboud University, Holland)**

**Title: **"Asymptotically mean value harmonic functions on nonsmooth spaces"

**Abstract: **A central property of harmonic functions in Euclidean domains is the mean value property. Although it is very specific to the Euclidean setting - and need not hold on more general Riemannian manifolds - the mv property can be stated using only the underlying metric and measure. In this talk I will introduce an extension of this, asymptotic mean value (amv) harmonicity, which can be defined on any metric measure space and which characterizes harmonicity in Riemannian manifolds. I will describe connections between harmonicity and amv harmonicity beyond the smooth setting, in Heisenberg groups and RCD spaces, and also present some regularity results for amv harmonic functions on general doubling metric measure spaces. This is joint work with T. Adamowicz and A. Kijowski.

**14-Apr: **no seminar, Great Thursday

**21-Apr: **no seminar, IM PAN's research council meeting

**28-Apr: M. Capolli (IM PAN)**

**Title: **"Legendrian Energy Minimizers in the Heisenberg Group"

**Abstract: **In their pioneering work, Korevaar and Schoen defined the Sobolev spaces of functions with target in a metric space. Among many results, they were able to obtain satisfactory existence and regularity results for energy minimizers under the assumptions that the Alexandrov curvature of the target space is non-positive. Some years later Capogna and Lin proved similar results in the Heisenberg group H^n, a metric space with unbounded Alexandrov curvature. They were able to obtain the existence for energy minimizers for any n and regularity for n=2. In this talk I will retrace the work of Capogna and Lin and I will explain how, together with Adamowicz and Warhurst, we are studying the extention of the regularity result to any n.

**5-May: G. Veronelli (University of Milano-Bicocca)**

**Title: **"The L^{p}-positivity preservation on Riemannian manifolds"

**Abstract: **A Riemannian manifold is L^{p}-positivity preserving if any L^{p} distributional solution of (−∆ + 1)u ≥ 0 is necessarily non-negative.In his seminal works on the spectral properties of Schrodinger operators with locally L^{p} potentials on the Euclidean spaces, T. Kato proved the self-adjointness of such operators by combining his celebrated inequality with the L^{p}-positivity preservation property of R^{n}. While trying to extend Kato’s investigations and techniques to Schrodinger operators on Riemannian manifolds, in 2002 M. Braverman, O. Milatovic and M. Shubin conjectured that any complete manifold is L^{2}-positivity preserving.

In this talk, we will first review the different approaches proposed to face this conjecture and the partial results obtained so far. Then we will present a new strategy based on regularity properties of subsolutions of elliptic PDEs, which permits to settle the problem. All the techniques will be mostly presented in the Euclidean setting, so as to avoid geometric technicalities as much as possible. It is a joint work with Stefano Pigola.

**12-May: Z. Grochulska (MIM UW)**

**Title: **"Homological and homotopical bounded turning properties in Euclidean spaces"

**Abstract: **Classical bounded turning property states that any two points $x, y$ of a domain $\Omega$ can be joined by a continuum whose diameter does not exceed $C |x - y|$ for some constant $C$. We will consider higher-dimensional versions of this condition (introduced by Alestalo and V\"{a}is\"{a}l\"{a} in the nineties) which can be phrased either in homological or homotopical manner. I will discuss an example of a domain for which the homological condition is satisfied and the homotopical is not and show the connection of these properties with the idea of John and uniform domains. The talk is based on the work in progress with P. Goldstein, C.-Y. Guo, P. Koskela and D. Nandi.

**19-May: M. Gryszówka (IM PAN)**

**Title: **"Epsilon-approximability in Lipschitz sets"

**Abstract: **Garnett proved that harmonic functions are epsilon-approximable in a half-plane by functions of bounded variations. However, just a couple of years later Dahlberg managed to extend that result to Lipschitz domains in R^n (AIF 1980). He used similar methods, but also introduced a few new elements to the proof. I will present a sketch of his proof with a special emphasis on those parts that are novelties delivered to prove a more general version. The epsilon-approximability is one of the key steps in the Corona theorems.

**26-May: B. Warhurst (MIM UW)**

**Title: **"L-harmonic contact symmetries of the Heisenberg Group"

**Abstract: **The basic idea of energy minimisation that leads to harmonic mappings between Riemannian manifolds can be considered in broader geometric settings. In this talk we take a look at these ideas on the subriemannian Heisenberg group in relation to L-harmonic contact transformations, where L is the sublaplacian.

**2-June: S. Golo (Jyvaskyla University, Finland)**

**Title: **

**Abstract:**

**9-June: M. Walicki (MiNI PW)**

**Title: **

**Abstract:**

FALL SEMESTER

**7-Oct: **Ben Warhurst (MIMUW)

__The seminar was cancelled and moved to 28.X__

**14-Oct: **Marco Capolli (IMPAN)

**Title: **A C^m Lusin Theorem in the Heisenberg Group

**Abstract: **In the Euclidean setting, a measurable function can be made continuous after a perturbation in a set of small measure. Concerning higher regularities it is possible to give conditions under which a function can be approximated by a function of class C^m. This results relies on the classical Whitney extension theorem. In this talk I will present joint work with Prof. Andrea Pinamonti (Trento, Italy) and Prof. Gareth Speight (Cincinnati, US) in which we studied an analogous result in the Heisenberg group.

**21-Oct: Rajchman Zygmund Marcinkiewicz conference, IMPAN 18-22 X. **

**No seminar.**

**28-Oct: **Ben Warhurst (MIMUW)

**Title: ** Schwarzians on the Heisenberg group

**Abstract: ** In the conformal mapping theory of the complex plane, the Schwarzian arises as the differential equation that characterises Möbius transformations. In this talk I will discuss attempts to define a Schwarzian on the Heisenberg group and the consequences of rigidity which.

**4-Nov: No seminar**

**11-Nov: The Independence Day.**

**No seminar.**

**18-Nov: **Zofia Grochulska (MIM UW)

**Title: **Non-classically differentiable homoemorphisms in R^n and their Jacobians

**Abstract: ** Homeomorphisms equipped with some kind of derivative are a natural choice for describing deformations of some materials. It is then interesting to understand the interplay between the analytical and topological properties such homeomorphisms have, for example if their Jacobians tell us if the mappings preserve or reverse orientation. I will present the example given by Goldstein and Hajłasz of an orientation preserving, approximately differentiable homeomorphism whose approximate Jacobian equals -1 almost everywhere and give an overview of related results concerning Sobolev homeomorphisms.

**25-Nov:**

**Title: **

**Abstract:**

**2-Dec:**

**Title: **TBA

**Abstract:**

**9-Dec: **Marcin Gryszówka (IM PAN)

**Title: TBA**

**Abstract:**

**16-Dec: **Michał Miśkiewicz (MIM UW)

**Title: TBA**

**Abstract:**

**23-Dec - 6-Jan : **Christmas holidays

**No seminar**

**13-Jan: **

**Title: **

**Abstract:**

**20-Jan: ** Marcin Walicki (MiNI PW)

**Title: ** "The geometry of level lines of solutions of uniformly elliptic equations in the plane"

**Abstract:**

**27-Jan: **

**Title: **

**Abstract:**

Academic year 2020/2021

The seminsar was suspended due to the Covid epidemic.

Academic year 2019/2020

**Fall **semester: the seminar was a part of the Simons semester in Geometry and Analysis:

https://www.impan.pl/en/activities/banach-center/conferences/19simons-xii/program

Spring semester: the seminsar was suspended due to the Covid epidemic.

Academic year 2018/2019

**8-Oct, DE Seminar: **Tomasz Cieślak (IM PAN),

**Title:** Multipeakons: between PDEs, geometry and control.

**18-Oct: **Antoni Kijowski (IMPAN)

**Title: ** "Well-posedness of PDEs on graphs"

**Abstract:** During my presentation I will discuss the question of well-posedness of PDEs on finite weighted graphs. I will introduce the notion of an (strictly) elliptic operator on graph and prove that for such class solutions are uniquely defined by the boundary data. One of the main tools is the comparison principle. Furthermore, I will show existence of solutions to homogeneous elliptic PDEs using the Brouwer fixed point theorem. The whole discussion will be illustrated by examples of PDE operators, e.g. the p-Laplacian for p \in [1,\infty] and the eikonal operator. Moreover, I will show examples of ill-posed PDEs on graphs, when either existence or uniqueness fails.

The talk is based on the paper by J. Manfredi, A. Oberman, A. Sviridov "Nonlinear elliptic partial differential equations and p-harmonic functions on graphs", Differential Integral Equations (2015).

**25-Oct: **Mark Pankov (UWM)

**Title: ** "Non-injective version of Wigner's theorem (from the space of pure states to Riemann sphere)"

**Abstract:** Wigner's theorem is one of basic results of mathematical foundations of quantum mechanics. It says that all symmetries of the space of pure states are induced by unitary and anti-unitary operators. Let $H$ be a complex Hilbert space and let $F_{s}(H)$ be the real vector space of all self-adjoint operators of finite rank. Denote by $P(H)$ the set of all rank one projections (the space of pure states). We describe all linear transformations of $F_{s}(H)$ sending rank one projections to rank one projections. The general problem can be reduced to the case when $\dim H=2$. In this case, $P(H)$ is identified with the Riemann sphere and linear transformations preserving rank one projections induce Mobius transformations of special type.

**1-Nov: **No seminar (All Saints Day).

**8-Nov: **Tomasz Kostrzewa (MiNI PW)

**Title:** "Harmonic functions and measures on trees"

**Abstract: **In my talk I will discuss different notions of harmonicity on (homogeneous) trees. In the second part of the talk I will focus on doubling measures on regular trees. In particular, I will construct a metric and a measure such that the considered tree is a doubling metric measure space.

**15-Nov: **No seminar

**19-Nov, DE Seminar: **Piotr Gwiazda (IMPAN), Title: "1/3-everywhere"

**22-Nov: **Sławomir Kolasiński (MIM UW)

**Title: **"Equivalence of ellipticity conditions for geometric variational problems"

**Abstract: **Ellipticity in geometric variational problems is a feature of a functional, defined on geometric objects (e.g. currents), which allows to prove existence and regularity of minimizers. The geometric objects we are dealing with are basically just d-dimensional subsets S of ℝⁿ and the functionals are defined by integrating certain integrand F over S with respect to the d-dimensional Hausdorff measure. The integrand may depend not only on the point in space but also on the tangent direction.

In my recent joint work (arXiv:1810.07262) with Antonio De Rosa we compare two different notions of ellipticity. The first one, denoted AE, was introduced in 1960s by Frederick Almgren who also laid the foundations of regularity theory for minimizers (Ann. of Math. 87, 1968). The second one, called AC, is a new definition that appeared in the work of De Philippis, De Rosa, and Ghiraldin (Comm. Pure Appl. Math. 71(6), 2018) as a sufficient and necessary condition for a minimizer (more generally, for any critical point) to be rectifiable.

It is very hard, in practice, to verify that a given functional satisfies AE. Actually, there are essentially no non-trivial examples (at least in case n-d > 1). The condition AC is easier, more algebraic and, due to the work of De Philippis, De Rosa, and Ghiraldin, we know it is exactly the right condition.

In our joint work with De Rosa we prove that AC implies AE. Interestingly, in the course of the proof we had to employ some (classical) algebraic topology. In contrast to my previous talk on this topic, I will focus on the proof rather than on the background.

**29-Nov: **Tomasz Cieślak (IMPAN),

**Title:** ""Almgren's frequency formula"

**Abstract**: In my talk I am going to introduce and prove Almgren's frequency formula for harmonic functions. This will require introducing harmonic polynomials and some facts

related to them.

**6-Dec: **Marta Szumańska (MIM UW),** **

**Title:** "Geodesic radius of curvature for horizontal curves in Heisenberg group**" **

**Abstract**: The intrinsic curvature of an Euclidean C^2 curve in Heisneberg group was introduced by Balogh, Tyson and Vecchi (It was obtained in a limiting process and is based on curvatures on Riemannian spaces approximating the Heisenberg group). For horizontal curves this curvature coincides with Euclidean curvature of its ortogonal projection onto XY-plane.

We define a notion of "global" curvature that can be considered for any horizontal curve (not necessarily C^2). The idea is based on the following fact: the image of the ortogonal projection into XY-plane of any geodesic in Heisenberg group is an arc of a circle. For any two points in Heisenberg group we define a geodesic radius of curvature which is the radius of the circle arc obtained by a the projection from the unique geodesic connecting those two points.

The aim of the talk is to show the similarities between the role played by the intrinsic curvature in Heisenberg group and "normal" curvature, and between the geodesic radius of the curvature and the Menger curvature in Euclidean space.

The presentation is based on a joint work in progress with Katrin Faessler.

** **

**10-Dec, DE Seminar: **Jarosław Mederski (IMPAN), TBA

**20-Dec: **No seminar

**9-Jan****, r. 405: **Mario Santilli (Augsburg University)

**Title**: "Rectifiability and approximate differentiability of higher order for sets"

**Abstract**: In this talk we introduce a novel concept of approximate differentiability of higher order for sets that allows to characterize C^{k,α}-rectifiable sets for every integer k≥1 and for every 0≤α≤1. This solves a problem left open from the work of Anzellotti and Serapioni on this subject. Moreover, our result naturally extends to the class of sets classical results for approximate

differentiable functions proved by Federer, Isakov and Whitney.

**23-Jan, r. 405****: **Antonio De Rosa (Courant Institute, NYU)

**Title**: "Anisotropic counterpart of Allard’s rectifiability theorem and Plateau problem"

**Abstract**: We present our extension of Allard's celebrated rectifiability theorem to the setting of varifolds with locally bounded anisotropic first variation. We identify a necessary and sufficient condition on the integrand for its validity and we discuss the connections of this condition to Almgren's ellipticity. We apply this result to the set-theoretic anisotropic Plateau problem, obtaining solutions to three different formulations: one introduced by Reifenberg, one proposed by Harrison and Pugh and another one studied by David. Moreover, we apply the rectifiability theorem to prove an anisotropic counterpart of Allard's compactness result for integral varifolds.

Some of the presented theorems are joint works with De Lellis, De Philippis, Ghiraldin and Kolasiński.

**31-Jan: **Maria J. Gonzalez (University of Cádiz)

**Title:** "Carleson measures on simply connected domains"

**Abstract**: Carleson characterized the positive measures on the unit disc for which the Hardy spaces embed continuously in L^p of the measure. This theorem has led

to various generalizations, including similar results for the weighted Bergman spaces. In this talk we will review these classical results and show how similar characterizations hold on more general domains of the plane.

**11-Mar, DE Seminar: **Wojciech Zajaczkowski (IMPAN),

**Title**: On free boundary problem for viscous incompressible magnetohydrodynamics (mhd)

**21-Mar****:**** **Antoni Kijowski (IMPAN)

**Title: **"Harmonic functions are locally Lipschitz on graphs"

**Abstract**: During my talk I am going to show that harmonic functions on infinite connected graphs with uniformly bounded vertex degree are locally Lipschitz. I will illustrate the discussion with an example of a harmonic function which local Lipschitz constant blows up.

The talk is based on a paper by Lin and Xi: "Lipschitz property of harmonic function on graphs", JMAA (2010).

**28-Mar: **Panayotis Smyrnelis (IMPAN)

**Title: "**Solutions of semilinear elliptic PDE with convex potentials"

**Abstract**: I will start with a description of the orbits of the ODE problem. Then, I will present two relevant results for the scalar PDE:

1) the existence of blowing up solutions (which has initially been studied by Loewner and Nirenberg), and

2) the convexity of solutions in convex planar domains with constant Dirichlet boundary condition (established by Caffarelli and Friedman).

Finally, I will discuss an interesting open problem: the validity of a Rado-Kneser-Choquet type theorem for the vector PDE.

**4-Apr: **Ben Warhurst (MIM UW)

**Title: "**Teichmuller Theory on the Heisenberg Group"

**Abstract**: The talk will discuss the Beltrami equation and the Schwarzian derivative on the Heisenberg group. The talk is based on the work in progress with Tomasz Adamowicz.

**8-Apr, DE Seminar: Agnieszka Świerczewska-Gwiazda (MIM UW)**

**Title: **Transport equation - from renormalizations to regular Lagrangian flows

**Abstract**: My talk will concentrate on various methods for transport equation with non-regular transport coefficient. A starting point will be the concept of renormalized solutions to transport equation introduced by R. DiPerna and P.-L. Lions. In the classical setting the connection to corresponding ordinary differnetial equation is straightforward. However extensions to less regular vector fields are of significant interest and motivations for such studies arise from many equations of mathematical physics. I will show an example of a transport equation with an integral term and discuss the problems of existence, uniqueness and stability. For this problem, which comes from description of polymeric flows, we will see that the approach of DiPerna and Lions is not efficient and I will explain how the approach of regular Lagrangian flows can be used in this case. The last part of the talk is a common result with Camillo De Lellis and Piotr Gwiazda.

**18-Apr: **No seminar (Good Thursday)

**25-Apr: **Michał Miśkiewicz (MIM UW)

**T****itle:** "Stability of singularities of minimizing harmonic maps "

**Abstract**: Minimizing harmonic maps - i.e., maps into a fixed manifold that minimize the Dirichlet energy - are known to be smooth outside a singular set of codimension 3. Here, we consider maps into the standard sphere S2 and investigate how the singular set is affected by small perturbations of the prescribed boundary map. We show a simple stability result in which the singularities of two minimizing maps are compared using the Wasserstein distance. The talk is based on joint work with Katarzyna Mazowiecka and Armin Schikorra.

**29-Apr: 11.30, r. 403, **Michael Cowling (UNSW Sydney, Australia)

**Title:** ``Connected locally compact homogeneous metric spaces"

**Abstract**: A metric space is homogeneous if its isometry group acts transitively. Recent work with Ville Kivioja, Enrico Le Donne,

Sebastiano Nicolussi Golo, and Alessandro Ottazzi, studies these spaces. In particular, we describe all such metric spaces ``to within

$\epsilon$'' as Lie groups with translation-invariant distance functions, and describe the metric spaces that arise as parabolic

boundaries of homogeneous Riemannian metric spaces of negative curvature. This talk reviews part of this work.

**9-May: **Sebastiano Golo (University of Padova)

**Title: ** "Spectral multipliers and wave equation for sub-Laplacians"

**Abstract**: Mihlin--Hörmander theorem gives the sharp Sobolev order $n/2$ for a spectral multiplier of the Laplacian to define a bounded operator on $L^p(\R^n)$ for all $p\in(1,\infty)$.

We study the same type of statements for sub-Laplacians, which are sub-elliptic operators defined on sub-Riemannian manifolds.

Although a Mihlin--Hörmander-type theorem in Carnot groups is known, the sharp Sobolev order is still unknown. It is conjectured to be $n/2$, where $n$ is the topological dimension.

We have proven that in no sub-Riemannian manifold the sharp Sobolev order can be lower than $n/2$, where $n$ is the topological dimension. For the proof, we construct a partial representation of the sub-Riemannian half-wave propagator as a Fourier integral operator. For such Fourier integral operator, the critical points of the phase function are determined by the sub-Riemannian exponential map.

This is a joint work with Alessio Martini and Detlef Müller.

**13-May,** **DE Seminar: **Andrzej Szulkin (Stockholm)

**Title: "**A simple variational approach to weakly coupled elliptic systems"

**30-May: **Antoni Kijowski (IM PAN)

**Title: ** "Asymptotic p-mean value property of p-harmonic functions"

**Abstract**: During my talk I will discuss the following result: for a continuous function in Euclidean domain the following conditions are equivalent: 1) to be a viscosity solution to the p-Laplace equation, and 2) to possess an asymptotic p-mean value property in the viscosity sense. I will explain the proof for p \in (1,\infty] and show why the proof does not work in the case p=1. Moreover, I will discuss a generalization to the setting of Carnot-Caratheodory groups. The Euclidean part of the talk is based on Ishiwata-Magnanini-Wadade, Calc. Var. (2017), the second part is based on joint work with Adamowicz, Pinamonti and Warhurst.

**6-June: **Jan Lang (Ohio State University)

**Title: ** "p-Laplacian and generalized trigonometric functions"

**Abstract**: Motto: "Etwas allgemein machen, heißt, es denken", W.F.Hegel (1833)

Generalized trigonometric functions, which were first introduced by Lundberg 1879, will be discussed together with their connections with p-Laplacian, Approximation theory and s-numbers for Sobolev embedding.

**10-June,** **DE Seminar: **Wojciech Kryński (IMPAN)

**Title: "**Cone structures on manifolds and PDEs"

Academic year 2017/2018

**5-Oct: **Sebastiano Golo (University of Jyvaskyla, Finland)

**Title**: Homogeneous distances and shape of spheres

**Abstract**: A graded groups are a generalization of Carnot groups introduced already by Folland and Stein in the Eighties. On graded groups it is possible to define canonical anisotropic dilations. The distances that are both left-invariant and 1-homogeneous with respect to these dilations are called "homogeneous distances". I will present some results on the regularity of spheres of homogeneous distances on graded group. As an application, I will state a theorem about the Besicovitch Covering Property in the Heisenberg group.

**12-Oct: **Sebastiano Golo (University of Jyvaskyla, Finland)

**Title**: Minimal surfaces in the Heisenberg group

**Abstract:** The Heisenberg group has a well studied sub-Riemannian structure. Geometric measure theory in the sub-Riemannian setting is still in development and several fundamental questions are still open. One reason is that sets of finite sub-Riemannian perimeter may have fractal behaviours. I will present the state of art (in my knowledge) of the study of minimal surfaces in this setting, with some recent results about contact variations of the area functional.

**19-Oct: **Michael Eastwood (University of Adelaide)

**Title**: Puncture repair in conformal geometry

**Abstract: **If you remove a point from a compact Riemann surface and ask for a conformal compactification of the result, then the only option is to replace the point. Roughly speaking, there is no difference between a punctured Riemann surface and a marked Riemann surface. The same result holds for conformal geometry in higher dimensions, where the proof should be easier but all the proofs I know are more difficult! There are plenty of ways to attempt to prove these results. I'll explain what I know. Sadly, after one successful argument, this will mostly be an explanation of what goes wrong or things I cannot prove. In other words, mostly sketches of various failed attempts. Needless to say, help from the audience would be much appreciated!

**26-Oct: **Michał Gaczkowski (IM PAN)

**Title**: Improved compactness in Hajłasz spaces

**Abstract: **On the Riemannian manifolds a compact Sobolev embedding can be improved in the presence of symmetries. We would like to investigate if such kind of results holds for Sobolev spaces defined on metric spaces. In this case most methods are very difficult to use. However, it turns out that using concentration compactness of P.L. Lions one can obtain better compact embeddings (unfortunately, not as good as expected). In my talk I would like to show how one can obtain such improvement in Hajłasz spaces.

**2-Nov: **Katrin Fässler (University of Fribourg, Switzerland)

**Title**: Singular integrals, intrinsic graphs and Lipschitz harmonic functions in the Heisenberg group

**Abstract: **The Heisenberg group H is a Lie group whose Lie algebra is given by the bracket condition [X,Y]=T. The differential operator X^2+Y^2 (sub-Laplacian) admits a fundamental solution G, which is closely related to a homogeneous norm on H. The resulting metric space (H,d) exhibits an interesting geometry. In this talk I will discuss the L^2 boundedness of certain 3-dimensional singular integral operators T with respect to 3-regular measures in (H,d). I will show that proving uniform L^2 boundedness of T on vertical planes -- which is essentially a Euclidean problem -- suffices to prove L^2 boundedness of T on intrinsic graphs of compactly supported intrinsic C^{1,\alpha} functions over vertical planes. This result applies to the operator T whose kernel is the gradient of G. As a corollary we infer that the intrinsic graphs mentioned above are non-removable for Lipschitz continuous solutions to the sub-Laplace equation. This is joint work with V. Chousionis and T. Orponen.

**9-Nov: **Antoni Kijowski (IMPAN)

**Title**: Characterization of strongly harmonic functions

**Abstract: **We say that a locally integrable function on a metric measure space is strongly harmonic if it attains the mean value property. In case of an open subset of Euclidean space with induced metric and measure it is equivalent, by the converse to the Gauss theorem, to solving the Laplace equation. However, in a general metric measure space X the class H of all such functions is not well characterized.

During my talk I will try to give the audience an insight into the structure and properties of H by investigating strongly harmonic functions on the Euclidean space equipped with a distance function and a weighted Lebesgue measure. I will show characterization of the class H in two cases: (1) when the distance is arbitrary and the measure is Lebesgue, and (2) with an arbitrary weight and the Euclidean distance. Finally, I will describe the dimension of space H.

**16-Nov: **No seminar. The conference "International Conference on Symmetry and Geometric Structures" at IM PAN

**23-Nov: **Iwona Skrzypczak (IMPAN)

**Title**: Lavrentiev's phenomenon in the Musielak-Orlicz spaces

**Abstract: ** Lavrentiev's phenomenon originally meant the situation when the infimum of the variational problem over the smooth functions is strictly greater than infimum taken over the set of all functions satisfying the same boundary conditions. The notion of the Lavrentiev phenomenon became naturally generalised to describe the situation, where functions from certain spaces cannot be approximated by regular ones (e.g. smooth).

Typical examples of the Musielak-Orlicz spaces are the variable exponent spaces, the Orlicz-Sobolev spaces, and the double-phase space.

It is well known that the smooth functions are dense in the variable exponent spaces only if the exponent is regular enough. Similar fact holds for the double-phase space. In the Orlicz-Sobolev spaces the result corresponding to the Meyers-Serrin theorem is Gossez's approximation theorem (Studia Math. '82), who infers that weak derivatives are strong derivatives with respect to the modular topology.

The talk will describe unification of these approaches provided in the manuscript arXiv:1711.06145: results on modular density of smooth functions. The precision of the method is confirmed by sharp results in all above mentioned cases. The possible applications to PDEs will be mentioned.

**30-Nov: **No seminar

**7-Dec: **Ben Warhurst (MIMUW)

**Title**: Puncture repair in metric measure spaces

**Abstract: **The talk will discuss a geometric measure space approach to the puncture repair problem in conformal Riemannian geometry discussed in the seminar earlier this semester by Mike Eastwood.

**14-Dec: **Maria J. Gonzalez (Universidad de Cádiz, Spain)

**Title**: Geometry of quasicircles and Beurling type operators

**Abstract: **The general problem is the following: Given a quasiconformal mapping ρ on the plane, what conditions on its dilatation µ guarantee certain geometric properties of the quasicircle Γ =ρ(R)? In this talk we will mainly consider the case where Γ is a chord-arc curve. Denoting by S the Beurling operator, we will show that the invertivility of the operator (I−μS) on a particular weighted L^2 space characterizes such curves. (Joint work with K. Astala)

**1-Mar: **Paweł Goldstein (MIMUW)

**Title**: Topologically non-trivial counterexamples to Sard's theorem

**Abstract: **In 1942, A. Sard proved his celebrated theorem: the set of critical values of a mapping between R^m and R^n is of measure zero, provided the mapping is sufficiently regular (C^k, for k>mam-n,0)). The theorem can be easily applied to mappings between closed manifolds, showing that if f\in C^k(M,N) is surjective and k>\dim M-\dim N\geq 0, then there is an open set \Omega in M such that rank Df=\dim N in \Omega and f(\Omega is dense in N.

A mapping f\in C^2(S^{n+1},S^n) which is not homotopic to a constant map (such mappings exist for every n\geq 2) is surjective, and thus, by Sard's theorem, the rank of Df is n a.e. in M.

Sard's theorem does not hold any more if k\leq \dim M-\dim N and in fact we can construct, mimicking an example due to Kaufman, a surjective mapping f\in C^1(S^{n+1},S^n) with rank Df\leq 1 everywhere. However, by construction, this mapping is homotopic to a constant map.

In joint work with Piotr Hajłasz and Pekka Pankka we address the following question: Assume f:S^{n+1}\to S^n is a C^1, surjective mapping that is not homotopic to a constant map. Can the rank of the derivative Df be non-maximal a.e. in S^{n+1}?

We arrive at a surprising dichotomy:

If n=2 or n=3, then the rank of Df must be maximal on a set A\subset S^{n+1} of positive measure and f(A) is dense in S^n,

However, for n=4,5,\ldots there exists a map f\in C^1(S^{n+1},S^n) with rank Df

To prove the first result (and a bit more), we use a generalization of the Hopf invariant due to Hajłasz, Schikorra and Tyson. The second one is an explicit example, modeled on a result of Wenger and Young.

**5-Mar (!Monday!): **Ben Warhurst (MIM UW)

**Title**: A Koebe distortion theorem for quasiconformal mappings in the Heisenberg group

**Abstract: **The talk will discuss recent work with Tomasz Adamowicz and Katrin Fässler where we prove a Koebe distortion theorem for the average derivative of a quasiconformal mapping between domains in the sub-Riemannian Heisenberg group H1.

**15-Mar: **No seminar

**20-Mar(!Tuesday!): **Marcin Preisner (University of Wroclaw)

**Title**: MULTIPLIER THEOREMS ON SPACES OF HOMOGENEOUS TYPE

**Abstract: **

**29-Mar, 5-Apr, 12-Apr: **No seminar

**19-Apr: **Jose Llorente (Universitat Autònoma de Barcelona)

**Title**: A nonlinear mean value property related to the p-laplacian

**Abstract: **The Mean Value Property for harmonic functions is the essential element of the interplay between Potential Theory, PDE's and Probability.

During the last ten years an increasing attention has been paid to figure out which are the natural stochastic models for certain

nonlinear PDE's like the p-laplacian or the infinity-laplacian and the key point is to determine first which are the corresponding (nonlinear)

mean value properties. After reviewing some classical facts, we will introduce several variants of a nonlinear mean value property

associated to the p-laplacian. Then we will give some recent results in the euclidean space and in the more general metric space setting.

**26-Apr:** Michał Miśkiewicz (MIM UW)

**Title**: Singularities of minimizing harmonic maps

**Abstract: **Minimizing harmonic maps between manifolds are known to be smooth outside the so-called singular set. In general this is a rectifable set of codimension 3, i.e., it can be covered by countably many Lipschitz pieces, but still may have many small gaps. In one special case of maps from a 4-dimensional domain into the 2-dimensional sphere Hardt and Lin proved that the singular set consists of topological curves. I will show a generalization to higher dimensional domains and discuss the topological obstruction responsible for preventing gaps in the singular set.

**3-May:** No seminar (national holidays)

**8-May (!Tuesday, r. 405):** Aleksandr Logunov (IAS Princeton)

**Title**: Zero sets of Laplace eigenfunctions

**Abstract:** Nadirashvili conjectured that for any non-constant harmonic function in R^3 its zero set has infinite area. Nadirashvili's conjecture is true and we will discuss it's applications to the Yau conjecture on zero sets of Laplace eigenfunctions. Both conjectures can be treated as an attempt to control the zero set of a solution of elliptic PDE in terms of growth of the solution.

**17-May:** No seminar

**24-May:** Sławomir Kolasiński (MIM UW)

**Title**: Ellipticity in geometric variational problems

**Abstract: **We consider integral functionals defined on geometric objects (like currents or varifolds) and strive to find sufficient conditions on the integrand so to ensure existence (and regularity) of minimisers in certain classes

of competitors. Almgren suggested the notion of ellipticity (referred to as (AE)) based on Morrey's quasi-convexity.

A rather easy computation shows that convex integrands acting on rectifiable currents are (semi)elliptic; thus, in case of currents we have a lot of examples. However, for varifolds it is extremely hard to verify ellipticity. Recently,

De Philippis, De Rosa, and Ghiraldin suggested a different condition (called (AC)) which is also sufficient for existence and rectifiability of minimisers. Actually, it is necessary for rectifiability of stationary points. In co-dimension one, it is also equivalent to convexity of the integrand.

In our ongoing joint work with Antonio De Rosa we define another condition (called (BC)) which is equivalent to (AC)

and we try to show (without success so far) that (BC) implies (AE).

**29-May (!Tuesday, 15.00, r. 403):** Silvia Ghinassi (Stony Brook University)

**Title**: Sufficient conditions for $C^{1,\alpha}$ parametrization and rectifiability

**Abstract: **We provide sufficient conditions, in terms of Peter Jones’ $\beta$ numbers, for a set in $\mathbb{R}^n$ to be parametrized by a $C^{1,\alpha}$ map. We use this to obtain sufficient conditions for a set or measure in $\mathbb{R}^n$ to be $C^{1,\alpha}$ $d$-rectifiable, with $\alpha \in [0,1]$. The conditions use a Bishop-Jones type square function and all statements are quantitative in that the $C^{1,\alpha}$ constants depend on such a function. Key tools for the proof come from Guy David and Tatiana Toro's parametrization of Reifenberg flat sets (with holes) in the H\"{o}lder and Lipschitz categories.

**7-June (MIMUW, 14.15, r. 4050):** Sauli Lindberg (IC MAT)

**Title**: Hardy space theory of compensated compactness quantities with applications to fluid dynamics

**Abstract: **The topic of the talk is the Hardy space theory of compensated compactness quantities which originated in the celebrated paper of Coifman, Lions, Meyer and Semmes from 1993. I discuss the following

question of Coifman & al. in the case of partial differential operators: is H^1 integrability, under natural conditions, equivalent to compensated compactness? I also discuss the use of Hardy space theory in fluid dynamics and in particular my recent joint work with Daniel Faraco on uniqueness of weak solutions with vanishing Cauchy

data in 2D magnetohydrodynamics.

**14-June:** Tomasz Kostrzewa (MiNI PW)

**Title**: TBA

**21-June:** Antoni Kijowski (IM PAN)

**Title**: TBA

**28-June:** No seminar

Academic year 2016/2017

**20-Oct: **Tomasz Adamowicz (IM PAN)

**Title**: "Prime ends and mappings"

**Abstract**: The studies of prime ends have long history involving various approaches, for example due to Caratheodory, Nakki, Vaisala and Zorich. In the talk we present some of those theories, including recent developments in metric measure spaces. Moreover, we discuss problems of continuous and homeomorphic extensions of mappings to the topological and prime end boundaries in the Euclidean setting and the setting of metric measure spaces for a class of mappings generalizing quasiconformal mappings.

**27-Oct:** Tomasz Adamowicz (IM PAN)

**Title**: "Prime ends and mappings, p. II"

**3-Nov: **Ben Warhurst (MIM UW)

**Title**: "Sub-Laplacians and Mean values"

**Abstract**: The talk will examine the relationship between the property of a function on a Carnot group being harmonic with respect to a sub-Laplacian, and the property of a function satisfying a mean value property.

**10-Nov: **Michał Miśkiewicz (MIM UW)

**Title**: "Rectifiable Reifenberg-type theorems"

**Abstract**: Reifenberg's topological disk theorem gives a criterion (the so-called Reifenberg flatness condition) on a closed subset S of R^n implying that S is bi-Holder with R^k. As shown by elementary examples, Reifenberg flat sets can have infinite k-dimensional measure and even dimension greater than k. There are several generalizations with stronger assumptions that yield an upper bound on the k-dimensional measure of S (T. Toro, G. David). In this talk I will discuss a recent result due to A. Naber and D. Valtorta, which has relatively weak assumptions and allows some interesting generalizations.

**17-Nov: **Antoni Kijowski (IM PAN)

**Title**: "Connections between harmonicity and mean value property"

**Abstract**: One of many properties of harmonic function *u* is that the mean value of u over any ball *B(x,r)* is equal to the *u(x)* (mean value property). The converse is also true, meaning that if a continous function u has mean value property on every ball in its domain, then u is harmonic. During my talk we will see how can we weaken the assumptions, to still obtain harmonicity, particularly considering lower regularity of function or smaller number of radii on which the mean value property holds (e.g. one radius and two radius theorems). A number of examples will help us understand which assumptions are essential.

**24-Nov: **Martyna Patera (IM PAN)

**Title**: "The Mean Value Property for Harmonic Functions on Trees"

**Abstract**: One of the most important properties of harmonic functions is the mean value property, which says that the average value of a harmonic function over any admissible ball is equal to the value of the function at the center of the ball. Conversely, functions having the mean value property are harmonic (under some assumptions).

The main part of my talk will be dedicated to one radius theorems in the discrete setting. For a function satisfying the mean value property for one radius (depending on the point), we will see what conditions on the radius function are needed for the function to be harmonic. We will consider harmonic functions in R^n and on trees. We will define them on trees in two ways: using the mean value property and using the discrete Laplacian. We will also investigate connections between these two definitions.

**1-Dec: **Jarosław Mederski (IM PAN & UMK)

**Title**: "Linear and semilinear curl-curl problems"

**Abstract**: We present recent results concerning linear and semilinear equations involving the curl-curl operator. Our problems are motivated by nonlinear materials and solutions to the curl-curl problems lead to time-harmonic electromagnetic waves propagating in the materials. We discuss different nonlinear materials e.g. with the optical Kerr effect.

**8-Dec: **No seminar: *Winter school in PDEs,* Będlewo (Simons Semester)

**15-Dec: **Jacek Gałęski (MIM UW)

**Title**: "Besicovitch-Federer projection theorem for mappings having constant rank of the Jacobian matrix"

**Abstract**: For an m-unrectifiable set S of finite Hausdorff measure and a mapping f from R^n to R^n having constant, equal to m, rank of the Jacobian matrix, we prove that in any open neighbourhood of f in C^1 topology there exist a mapping g such that Hausdorff measure of the image g(S) is zero. Local (restriction to a small set in the domain) version of the above theorem is an easy consequence of the Besicovitch-Federer projection theorem. The main body of the talk is the construction of one mapping from parts (mapping defined on small open domain) that do not agree with other parts on intersections of their individual domains.

**22-Dec, 29-Dec, 5-Jan: **No seminars, Christmas Holidays

**12-Jan: **Tomasz Kostrzewa (MiNI PW)

**Title**: "Differential operators on LCA groups"

**Abstract**: In my talk I will introduce the notion of an invariant differential operator on locally compact abelian group and study its properties and relations to the Bruhat-Schwartz class. I will also talk about my attempts on using this notion in the theory of Sobolev spaces on LCA groups.

**19-Jan: **Marta Szumańska (IM PAN & MIM UW)

**Title**: "Gol'dshtein and Ukhlov results on traces of functions of Dirichlet spaces defined on simply connected domains"

**Abstract**:The Dirichlet space is a space of locally integrable functions whose weak gradient is square integrable on a given domain. Such functions are defined up to the set of measure zero, thus their values on the boundary of the domain can be considered only in terms of trace. Gol'dshtein and Ukhlov proved that if the Dirichlet space is defined on a simply connected domain in the plane, connected at each boundary point, then the unique quasi-continuous representative of a function in the Dirichlet space can be extended to quasi-continuous function on the closure of the domain; the extended function is defined on the boundary up to the set of capacity zero.

In case of broader class of domains, i.e. simply connected domains, one can construct the extension to the Caratheodory boundary (up to the set of capacity zero).

During the talk I am planning to show main ideas of the proofs of the abovementioned theorems and introduce necessary tools and notions (properties of the Dirichlet space and capacities, notions of quasi-continuous functions, the Caratheodory boundary and the capacitary boundary).

The talk is based on the paper ,,Traces of functions of L_2^1 Dirichlet spaces on the Caratheodory boundary" by V. Gol'dshtein A. Ukhlov; Studia Mathematica 235 (2016), 209-224.

**26-Jan: **Stanislav Hencl (Charles University, Prague), room **321**

**Title**: "Jacobians of Sobolev homeomorphisms"

**Abstract**: We show that if $n$ is 2 or 3, $f$ is a homeomorphism of a domain $\Omega$ in $R^n$ and $f$ is in the Sobolev space $W^{1,1}_{\loc}(\Omega,R^n)$, then the Jacobian determinan $J_f$ is either non-negative or non-positive a.e. in $\Omega$. This answers an open problem by P. Hajlasz. It is a joint work with J. Maly.

**2-Feb: **Teri Soultanis (IM PAN), room 403

**Title**: "Energy minimization of maps between metric spaces, and hyperbolicity in the target"

**Abstract**: I present the class of Newtonian maps between metric spaces and discuss the notion of homotopy of such maps. The role of nonpositive curvature in the target for proving existence of energy minimizers in homotopy classes is well known but in this generality the proof seems to require an extra hyperbolicity assumption. I will explain the main ideas in the proof and illustrate where the hyperbolicity comes in.

**2-Mar: **Yevhen Sevostianov (Zhytomyr State University), r. 403

**Title**: "Geometric Approach in the Theory of Spatial Mappings"

**Abstract**: The talk is devoted to the investigation of the space mappings with non-bounded characteristics of quasiconformality. In particular, we mean here the so-called mappings with finite distortion which are intensively investigated by leading mathematicians in the last decade. The series of properties of the so-called ring Q–mappings are obtained. The above mappings include the mappings with bounded distortion by Reshetnyak (quasiregular mappings). In particular, the differentiability and the analogues of the theorems of Sokhotski–Weierstrass, Liouville, Picard, Iversen, Montel etc. are obtained for the above mappings.

**9-Mar: **Yevhen Sevostianov (Zhytomyr State University), r. 403

**Title**: "On boundary behavior of mappings in terms of prime ends"

**Abstract**: A boundary behavior of mappings, which are closely related with the Sobolev and Orlicz--Sobolev classes in the plane and in the space, is investigated. There are obtained theorems on the boundary behavior of classes mentioned above.

**16-Mar: **No seminar. "Workshop on Nonstandard Growth Analysis and its Applications 2017"

http://crossfields.impan.pl/growth.html

**23-Mar: **Michał Miśkiewicz (MIM UW), r. 403

**Title**: "Fractional differentiability of p-harmonic functions"

**Abstract**: It is well known that the gradient of any p-harmonic function is locally Holder continuous (with some small exponent α dependent on p and the dimension) - this can be viewed as existence of derivatives of order α in terms of Besov spaces. My talk is concerned with existence of higher order derivatives: for p>2 of order 2/p (G. Mingione) and for 2

**30-Mar: **Michał Gaczkowski (IM PAN & MiNI PW), r. 403

**Title**: "Concentration compactness in variable exponent spaces"

**Abstract**: When dealing with PDEs, compact embeddings are one of the most powerful tools. Unfortunately, they are no always valid. One can try to deal with this by seeking convergence in some other sense. In my talk I would like to introduce results of Fernandez and Silva about concentration compactness of P.L. Lions for variable exponent in $R^n$. I will also discuss how such kind of results could be possibly use to improve compact embeddings.

**6-Apr: **Sławomir Kolasiński (MIM UW), r. 403

**Title**: ''Solution of an anisotropic inhomogeneous Plateau problem''

**Abstract**: The Plateau problem is about finding a "surface" which minimizes "area" amongst competitors which "span" a given boundary. I shall briefly describe various formulations of the problem and shortcomings of the solutions. Next, I will present a sketch of the existence proof for an abstractly formulated problem destiled from Almgren's 1968 paper.

This talk will be based on my joint work with Yangqin Fang from the Max-Planck Institute in Potsdam.

**13-Apr:** No seminar (Good Thursday)

**20-Apr: **Martyna Patera (IM PAN), r. **408**

"Doubling measures on regular trees"

**Abstract**:

A measure on a metric space is a doubling measure if the measure of any ball

with radius 2r is bounded (with some multiplicative constant) by the measure of

a ball with radius r. A metric space equipped with a doubling measure holds

many useful properties.

The main part of my talk will be dedicated to studying a doubling condition on a

tree. On a regular rooted tree we can definie a metric and a measure to

obtain a metric measure space. We will investigate what properties a

radial measure, i.e. dependent only on the distance from the root, must

posses to be a doubling measure. In particular, we will look at a

uniformizing metric and the weighted measure associated with it.

**27-Apr: **Olli Toivanen (IM PAN), r. 403

**Title**: Regularity in generalized Orlicz spaces

**Abstract**: A generalized Orlicz space is a generalization of, for example, Orlicz spaces and Lebesgue spaces, both classical and of variable exponent. To put it in the terms of the integrability condition for a function t and a location x, with some exponent (function) p:

Lebesgue = t^p

Orlicz = \phi(t)

variable Lebesgue = t^{p(x)}

generalized Orlicz = \phi(x,t)

Recent research has began to look for the generalization of the various regularity theories of these cases.

I will discuss recent results by Harjulehto, Hästö, Klen, Cruz-Uribe, myself and others, starting from "Hölder regularity of quasiminimizers under generalized growth conditions" (Harjulehto, Hästö, T.; Calc. Var. PDEs, 56 (2), 2017.)

**4-May: **Antoni Kijowski (IM PAN), r. 106

**Title**: Regularity of weakly and strongly harmonic functions

**Abstract**: A strongly harmonic function is a function defined on metric measure space, having the mean value property for all balls. Weakly harmonic function are defined in a similar way, i.e. it has the mean value property at every point for at least one radius of a ball. During my talk I will discuss some consequences of endowing a distance function and a measure with additional properties on such type of harmonicity, i.e. the Hölder and Lipschitz continuity. I will present some examples, among them strongly harmonic functions which are spherically symmetric.

The talk is based on paper "Harmonic functions on metric measure spaces", T. Adamowicz, M. Gaczkowski, P. Górka, arXiv:1601.03919 and a work with T. Adamowicz.

**11-May: **Iwona Skrzypczak (IMPAN)

**Title**: Renormalized solutions to nonlinear elliptic partial differential equations

**Abstract**: The talk will concern existence of renormalized solutions to general nonlinear elliptic equation in Musielak-Orlicz space. The growth of

the leading part of the operator $A$ is controlled by a generalized nonhomogeneous and anisotropic $N$-function $M$. The approach does

not require any particular type of growth condition of $M$ or its conjugate $M^*$ (neither $\Delta_2$, nor $\nabla_2$ conditions).

The condition we impose is log-H\"older continuity of $M$, which results in good approximation properties of the space. The proof of the main results

uses truncation ideas, the Young measures methods and monotonicity arguments.

The results are the based on the joint work with Piotr Gwiazda and Anna Zatorska-Goldstein.

**18-May: **Ben Warhurst (MIM UW)

**Title**: Geometric function theory on Jets

**Abstract**: Jet spaces can be viewed as a generalisation of the Heisenberg group, particularly from a sub-Riemannian perspective. The talk will introduce jets as sub-Riemannian spaces and discuss some exemplary results.

**25-May: **Grzegorz Łysik (IM PAN & UJK)

**Title**: Twierdzenie o jednoznaczności dla funkcji $\rho$-analitycznych

**Abstract**: Zacznę od podania warunków koniecznych i dostatecznych na zachodzenie równości w nierówności trójkąta dla metryki niezmienniczej $\rho$ na $R^n$. Następnie zdefiniuję funkcje $\rho$-analityczne i pokażę, że posiadają one własność jednoznaczności, tzn. jeśli funkcja znika na zbiorze otwartym, to znika wszędzie.

**1-June: **Ben Warhurst (MIM UW)

**Title**: Geometric function theory on Jets, p.II

**Abstract**: Jet spaces can be viewed as a generalisation of the Heisenberg group, particularly from a sub-Riemannian perspective. The talk will introduce jets as sub-Riemannian spaces and discuss some exemplary results.