var sms = '<div id="sms_bulletin">  <div id="sms_header">  <div class="do_not_print" id="arrow_left"><a href="?next=-43"><img src="//www2.math.ethz.ch/smsjs/images/arrow_left.png"/></a></div><div class="do_not_print" id="arrow_right"><a href="?next=-41"><img src="//www2.math.ethz.ch/smsjs/images/arrow_right.png"/></a></div>    <div class="sms_title">      <span style="float: left;">SMS</span>      <span>SWISS MATHEMATICAL SOCIETY</span>      <span style="float: right;">SMG</span>    </div>    <div class="sms_info">      <span style="float: left; width: 250px;">Bulletin no. 2</span>      <span style="float: right; text-align:right; width: 250px;">25.9 - 29.9.2023</span>      <span style="text-align: center; display: block; width: 250px; margin: auto; position: relative;">Autumn Semester 2023</span>    </div>  </div>  <div></div>  <div id="sms_content">    <div class="day">      <div class="day_title">        Monday, September 25      </div>      <div class="day_content">        <div class="day_content_item normal"><div class="event_location uni_la"><a class="external" href="https://www.epfl.ch">EPF Lausanne</a></div><div class="event_type"><a class="external" href="https://www.unil.ch/dsa/en/home/menuinst/seminaires--evenements/research-seminars.html">Talks in Actuarial Mathematics </a></div><br /><div class="event_speaker"> Lanpeng Ji (University of Leeds, UK)</div><div class="event_title">Bayesian CART for insurance pricing</div><div class="event_time">14:00 &bull; EPF Lausanne, UniL campus, Extranef - 125</div>        </div>        <div class="day_content_item normal"><div class="event_location uni_zh"><a class="external" href="https://uzh.ch">Universit&auml;t Z&uuml;rich</a></div><div class="event_type"><a class="external" href="https://www.math.uzh.ch/mat760">Ergodic theory and dynamical systems seminar</a></div><br /><div class="event_speaker">Prof. Dr. Jayadev Athreya (University of Washington)</div><div class="event_title">Counting Pairs of Saddle Connections <span class="abstract-button do_not_print">abstract</span></div><div class="abstract-text" style="display: none"><strong>Abstract:</strong><br />Motivated by the comparison of sets of holonomies of saddle connections on translation surfaces to Poisson point processes, we show that for almost every translation surface the number of pairs of saddle connections with bounded virtual area has asymptotic quadratic growth. The proof techniques combine ergodic methods for counting saddle connections with the fact that the Siegel--Veech transform is in L2. The talk will not assume prior knowledge of translation surfaces, and we will attempt to keep it broadly accessible. This is joint work with Sam Fairchild and Howard Masur.</div><div class="event_time">15:00 &bull; UZH Irchel, Winterthurerstrasse 190, Z&uuml;rich, Building Y27, Room H 25</div>        </div>        <div class="day_content_item normal"><div class="event_location uni_be"><a class="external" href="https://www.unibe.ch">Universität Bern</a></div><div class="event_type"><a class="external" href="https://www.math.unibe.ch/research/talks/colloquium/index_eng.html">Mathematisches Kolloquium </a></div><br /><div class="event_speaker">Dr. Ovidiu Preda (Romanian Academy)</div><div class="event_title"><a class="external" href="https://www.math.unibe.ch/research/talks/colloquium/index_eng.html">Kähler and locally conformally Kähler metrics</a> <span class="abstract-button do_not_print">abstract</span></div><div class="abstract-text" style="display: none"><strong>Abstract:</strong><br />In many problems arising in differential geometry, an essential step is choosing a “good” metric. In complex geometry, we search for “good” metrics in the class of Hermitian metrics, which are Riemannian metrics compatible with the complex structure. In this class of Hermitian metrics, probably those which have the best properties are Kähler metrics. However, the existence of Kähler metrics on compact complex manifolds has a number of topological obstructions. Therefore, in non-Kähler geometry, one seeks to replace them with a suitable subclass of Hermitian metrics. One such subclass is represented by locally conformally Kähler (lcK) metrics, which were introduced by Vaisman. We present some properties of Kähler and lcK metrics on compact complex manifolds and then show how we can extend all the definitions to complex analytic spaces, and which results remain true in this more general setting.</div><div class="event_time">17:15 &bull; Universität Bern, Sidlerstrasse 5, 3012 Bern, Lecture Room B6</div>        </div>      </div>    </div>    <div class="day">      <div class="day_title">        Tuesday, September 26      </div>      <div class="day_content">        <div class="day_content_item normal"><div class="event_location uni_ge"><a class="external" href="https://unige.ch">Université de Genève</a></div><div class="event_type"><a class="external" href="https://www.unige.ch/math/annonces/groupes-geometrie">Groupes et Géométrie </a></div><br /><div class="event_speaker"> Jean-Pierre Eckmann (UNIGE)</div><div class="event_title">Tumbling Downhill along a Given Curve <span class="abstract-button do_not_print">abstract</span></div><div class="abstract-text" style="display: none"><strong>Abstract:</strong><br />A cylinder will roll down an inclined plane in a straight line. A cone will wigglealong a circle on that plane and then will stop rolling. We ask the inversequestion: For which curves drawn on the inclined plane $\\real^2$ can onechisel a shape that will roll downhill following precisely this prescribed curveand its translationally repeated copies?This is a nice, and easy to understand problem, but the solution is quiteinteresting.(Based on work mostly with Y. Sobolev and T. Tlusty. After a Nature paper,Solid-body trajectoids shaped to roll along desired pathways, August 2023).</div><div class="event_time">10:30 &bull; Université de Genève, Conseil Général 7-9, Room 1-05</div>        </div>        <div class="day_content_item normal"><div class="event_location uni_ba"><a class="external" href="https://unibas.ch">Universität Basel</a></div><div class="event_type"><a class="external" href="https://algebra.dmi.unibas.ch/seminar.html">Seminar Algebra and Geometry</a></div><br /><div class="event_speaker"> Pascale Voegtli (UCL London)</div><div class="event_title">Flop connections between minimal models for corank 1 foliations over threefolds <span class="abstract-button do_not_print">abstract</span></div><div class="abstract-text" style="display: none"><strong>Abstract:</strong><br />In recent years the understanding of the fundamental birational geometry of foliations, especially on 3-folds, has been promoted by several groundbreaking works. Cascini and Spicer have extended most parts of the classical MMP to threefold pairs equipped with a mildly singular corank 1 foliation. In particular, the existence of log-flips has been demonstrated. The successful establishment of a foliated analogue of the classical MMP in low dimensions naturally raises the question whether classical results being closely related to the MMP do find their natural generalizations to foliated pairs. One such classical result one might strive to convey to foliations is the well-known theorem of Kawamata from 2007, stating that two minimal models with terminal singularities are related by a sequence of flops. In the talk, after having introduced the relevant notions, we will sketch a proof of an analogue of Kawamata\'s theorem for foliated 3-folds. This is recent joint work with D. Jiao.</div><div class="event_time">10:30 &bull; Universität Basel, Spiegelgasse 5, Seminarraum 05.001</div>        </div>        <div class="day_content_item normal"><div class="event_location uni_la"><a class="external" href="https://www.epfl.ch">EPF Lausanne</a></div><div class="event_type"><a class="external" href="https://www.unil.ch/dsa/en/home/menuinst/seminaires--evenements/research-seminars.html">Talks in Actuarial Mathematics </a></div><br /><div class="event_speaker"> Evarist J.S. Stoja (University of Bristol, UK)</div><div class="event_title">The Taxonomy of Tail Risk</div><div class="event_time">11:00 &bull; EPF Lausanne, UniL campus, Extranef</div>        </div>        <div class="day_content_item normal"><div class="event_location uni_ge"><a class="external" href="https://unige.ch">Université de Genève</a></div><div class="event_type"><a class="external" href="https://www.unige.ch/math/annonces/seminaire-danalyse-numerique">Séminaire d\'Analyse Numérique </a></div><br /><div class="event_speaker"> Teilo Wahl (Unversité de Genève)</div><div class="event_title">Sur les potentielles limitations du Deep Learning et ses utilisations pour la résolution d\'EDPs <span class="abstract-button do_not_print">abstract</span></div><div class="abstract-text" style="display: none"><strong>Abstract:</strong><br />La première partie de cette présentation est consacrée aux résultats de l\'article "The troublesome kernel", qui montrent l\'instabilité des réseaux Deep Learning comparés aux méthodes classiques dans le cadre des problèmes de reconstruction d\'images à partir de données limitées. La seconde partie est centrée autour de l\'élaboration de réseaux neuronaux pour la résolution d\'EDPs, avec essais numériques effectués sur l\'équation de Poisson.</div><div class="event_time">14:00 &bull; Université de Genève, Conseil Général 7-9, Room 1-05</div>        </div>        <div class="day_content_item normal"><div class="event_location uni_la"><a class="external" href="https://www.epfl.ch">EPF Lausanne</a></div><div class="event_type"><a class="external" href="https://calendar.google.com/calendar/embed?mode=AGENDA&height=600&wkst=2&bgcolor=%23FFFFFF&src=t36it03v52cus2detrvinlgk18%40group.calendar.google.com&color=%235F6B02&ctz=Europe%2FZurich">Groups, Arithmetic and Algebraic Geometry Seminar </a></div><br /><div class="event_speaker"> Dimitar Jetchev</div><div class="event_title">The Birch and Swinnerton-Dyer formula in analytic ranks zero and one for modular forms of higher weight  <span class="abstract-button do_not_print">abstract</span></div><div class="abstract-text" style="display: none"><strong>Abstract:</strong><br />In this talk, I will report on recent results on the computation of the p-part of the leading term of the L-function of a modular form of arbitrary weight at the central point in the cases when the order of vanishing is at most one. Unlike the classical case of weight 2 modular forms, qualitatively different arguments are needed in the higher-weight case. After explaining the difference, I will indicate how one can use level-raising and (non-ordinary) p-adic deformations together with some of the arguments in weight 2 to obtain results in the case of general weights. This is joint work with Chris Skinner and Xin Wan.</div><div class="event_time">14:15 &bull; EPF Lausanne, MA A1 12</div>        </div>        <div class="day_content_item normal"><div class="event_location uni_zh"><a class="external" href="https://ethz.ch">ETH Z&uuml;rich</a></div><div class="event_type"><a class="external" href="https://math.ethz.ch/s/analysis-seminar">Analysis Seminar</a></div><br /><div class="event_speaker">Dr. Stefano  Decio (Institute for Advanced Study, Princeton)</div><div class="event_title">Zeros of Steklov eigenfunctions <span class="abstract-button do_not_print">abstract</span></div><div class="abstract-text" style="display: none"><strong>Abstract:</strong><br />A Steklov eigenfunction in a bounded domain is a harmonic function whose normal derivative at the boundary is proportional to the function itself. I will tell you most of what I know about the zero sets of such functions. A nice fact is that there are many zeros near the boundary: I will give a gentle proof of this in the first part of the talk. In the second, perhaps a little less gentle, part I will discuss some lower and upper bounds for the Hausdorff measure of the zero set; several questions remain unanswered. Comparisons with the (slightly) better understood case of eigenfunctions of the Laplace-Beltrami operator will also be provided.</div><div class="event_time">15:15 &bull; ETH Zentrum, R&auml;mistrasse 101, Z&uuml;rich, Building HG, Room G 43</div>        </div>        <div class="day_content_item normal"><div class="event_location uni_zh"><a class="external" href="https://uzh.ch">Universit&auml;t Z&uuml;rich</a></div><div class="event_type"><a class="external" href="https://www.math.uzh.ch/mat075.1">Zurich Graduate Colloquium</a></div><br /><div class="event_speaker">Filippo Fila-Robattino (Universität Zürich)</div><div class="event_title">What is... the Reduced Phase Space of (N=1,D=4) Supergravity? <span class="abstract-button do_not_print">abstract</span></div><div class="abstract-text" style="display: none"><strong>Abstract:</strong><br />In field theory, from a local Lagrangian it is possible to derive the phase space of the system. Roughly speaking, it represents the space of Cauchy data of the theory. Such space is equipped with a closed 2-form, whose kernel is often given by Hamiltonian vector fields generating (gauge) symmetries. The reduced phase space (RPS) is obtained as a symplectic manifold via quotienting the original phase space. In this talk, after a brief introduction and an example of RPS in a simple case, I will present the derivation (using a construction due to Kijowski and Tulcjiew) of the RPS of N=1 supergravity in dimension 4. If time allows, the corresponding BFV description is then carried out.</div><div class="event_time">16:00 &bull; UZH Zentrum, Building KO2, Room F 150</div>        </div>      </div>    </div>    <div class="day">      <div class="day_title">        Wednesday, September 27      </div>      <div class="day_content">        <div class="day_content_item normal"><div class="event_location uni_la"><a class="external" href="https://www.epfl.ch">EPF Lausanne</a></div><div class="event_type"><a class="external" href="https://calendar.google.com/calendar/embed?mode=AGENDA&height=600&wkst=2&bgcolor=%23FFFFFF&src=t36it03v52cus2detrvinlgk18%40group.calendar.google.com&color=%235F6B02&ctz=Europe%2FZurich">Groups, Arithmetic and Algebraic Geometry Seminar </a></div><br /><div class="event_speaker"> Eric Chen</div><div class="event_title">Duality of automorphic periods <span class="abstract-button do_not_print">abstract</span></div><div class="abstract-text" style="display: none"><strong>Abstract:</strong><br />The study of automorphic integrals has a long history which began with Hecke\'s Mellin transform of a modular form, and more recently it has provided an indispensable language with which to describe functoriality phenomena in the Langlands program. In the first half of this presentation, I will explain these developments via emblematic examples while rephrasing them in terms of the recent framework of relative Langlands duality proposed by Ben-Zvi--Sakellaridis--Venkatesh. In the second half, I will present joint work with Venkatesh in which we establish relative duality in certain "singular" examples with the aid of new numerical invariants of Galois representations that we call "nonabelian L-functions".  </div><div class="event_time">10:00 &bull; EPF Lausanne, </div>        </div>        <div class="day_content_item normal"><div class="event_location uni_fr"><a class="external" href="https://www.unifr.ch">Université de Fribourg</a></div><div class="event_type"><a class="external" href="https://commonweb.unifr.ch/_Science/Math/Pub/Math/seminars/OSgeom-top.html">Oberseminar Geometrie</a></div><br /><div class="event_speaker"> Gonzalo Ruiz Stolowicz (EFPL)</div><div class="event_title">Representations on infinite-dimensional hyperbolic spaces <span class="abstract-button do_not_print">abstract</span></div><div class="abstract-text" style="display: none"><strong>Abstract:</strong><br />The infinite-dimensional hyperbolic spaces are defined and some generalities of the  group representations in their isometry groups will be presented. The functions of complex hyperbolic type defined by Monod are a fundamental tool when studying infinite-dimensional representations.  These functions play a role with respect to hyperbolic representations analogous to that of positive type functions with respect to unitary representations.  With this tool, the representations of the groups PU(1,n) and PO(1,n) will be addressed. </div><div class="event_time">10:20 &bull; Université de Fribourg, 0.05 PER23</div>        </div>        <div class="day_content_item normal"><div class="event_location uni_zh"><a class="external" href="https://ethz.ch">ETH Z&uuml;rich</a></div><div class="event_type"><a class="external" href="https://math.ethz.ch/s/algebraic-geometry-and-moduli-seminar">Algebraic Geometry and Moduli Seminar</a></div><br /><div class="event_speaker">Prof. Dr. Motohico Mulase (UC Davis)</div><div class="event_title">À la recherche de courbes cachées <span class="abstract-button do_not_print">abstract</span></div><div class="abstract-text" style="display: none"><strong>Abstract:</strong><br />In some problems, sometimes finding a hidden curve behind the scene becomes a key to solve the problem. I will explain a few results from the past with a new perspective: (1) Construction of a D-module on a Jacobian; and (2) Finding the right coordinate on the hidden curve to understand the relation between enumeration problems and the quantities on the Deligne-Mumford moduli stack of curves. Then I will try to explain (3) An attempt toward describing the character variety on a curve in a combinatorial manner. The last one is an ongoing project with Olivia Dumitrescu (UNC).</div><div class="event_time">14:00 &bull; ETH Zentrum, R&auml;mistrasse 101, Z&uuml;rich, Building HG, Room G 43</div>        </div>        <div class="day_content_item normal"><div class="event_location uni_ba"><a class="external" href="https://unibas.ch">Universität Basel</a></div><div class="event_type"><a class="external" href="https://dmi.unibas.ch/de/forschung/mathematik/analysis/analysis-seminar">Analysis Seminar</a></div><br /><div class="event_speaker"> Bian Wu (MPI Leipzig)</div><div class="event_title">Scale invariant bounds for mixing in the Rayleigh-Taylor instability</div><div class="event_time">14:15 &bull; Universität Basel, Spiegelgasse 5, Seminarraum 05.002</div>        </div>        <div class="day_content_item normal"><div class="event_location uni_zh"><a class="external" href="https://uzh.ch">Universit&auml;t Z&uuml;rich</a></div><div class="event_type"><a class="external" href="https://www.math.uzh.ch/aa/?events">Neuchatel - St.Gallen - Zurich Seminar in Coding Theory and Cryptography</a></div><br /><div class="event_speaker">Tatjana Bossalini (University of Zurich)</div><div class="event_title">The ROCA Vulnerability: A Study on Coppersmith&#039;s Algorithm and its Applications <span class="abstract-button do_not_print">abstract</span></div><div class="abstract-text" style="display: none"><strong>Abstract:</strong><br />The search for strong security and effective algorithms is never-ending in the field of modern cryptosystems. Cryptography is the foundation of trust in the digital era, covering everything from our private communications to financial transactions and essential infrastructure. It enables us to share sensitive information and communicate with confidence, knowing that our data remains confidential and secure. This thesis presents an introduction to Coppersmith\'s algorithm. We start by laying the theoretical groundwork for three variants: univariate modular, bivariate integer, and multivariate modular case. Subsequently, we proceed to analyze the diverse range of applications that can be derived from these findings, including but not limited to the use of stereotyped messages, random padding for two messages, factoring with incomplete information, and the Chinese Remaindering with Errors problem. We developed and implemented code examples for each scenario, providing a tangible demonstration of the algorithm\'s efficiency. Finally, we discuss the ROCA vulnerability that was found in Estonia and its effects on the nation. This weakness allows an attacker to factorize the RSA modulus and compromise the security of the affected systems. The Estonian example emphasizes the significance of using secure implementation techniques, being careful when choosing random number generators, and being cautious when picking cryptographic libraries.</div><div class="event_time">15:15 &bull; UZH Irchel, Winterthurerstrasse 190, Z&uuml;rich, Building Y27, Room H 25</div>        </div>        <div class="day_content_item normal"><div class="event_location uni_zh"><a class="external" href="https://ethz.ch">ETH Z&uuml;rich</a></div><div class="event_type"><a class="external" href="https://math.ethz.ch/s/geometry-seminar">Geometry Seminar</a></div><br /><div class="event_speaker"> Simon Machado (EHTZ)</div><div class="event_title">Approximate subgroups and bounded cohomology <span class="abstract-button do_not_print">abstract</span></div><div class="abstract-text" style="display: none"><strong>Abstract:</strong><br />Approximate subgroups are subsets of groups that are stable under multiplication up to a finite multiplicative error. The study of approximate subgroups has long been motivated by problems regarding random walks,  expansion and growth in groups. A notion of regularity - coined laminarity - has plaid a crucial role in the understanding of their structure. In a recent breakthrough, Hrushovski noticed that failure of laminarity could be witnessed by a certain bounded cohomology class. I will explain why another notion of cohomology - that sits halfway between bounded cohomology and group cohomology - is more suitable. I will then discuss how this can be used to classify approximate lattices - a class of approximate subgroups first studied by Yves Meyer that generalises at the same time lattices of locally compact groups, Pisot numbers of a number field and Penrose tilings.</div><div class="event_time">15:45 &bull; ETH Zentrum, R&auml;mistrasse 101, Z&uuml;rich, Building HG, Room G 43</div>        </div>        <div class="day_content_item normal"><div class="event_location uni_la"><a class="external" href="https://www.epfl.ch">EPF Lausanne</a></div><div class="event_type"><a class="external" href="https://memento.epfl.ch/event/probability-and-stochastic-analysis-seminar/">Probability (and Stochastic Analysis) Seminar </a></div><br /><div class="event_speaker"> Yatin Dandi (Bernoulli Centre)</div><div class="event_title">Title T.B.A.</div><div class="event_time">16:00 &bull; EPF Lausanne</div>        </div>        <div class="day_content_item normal"><div class="event_location uni_zh"><a class="external" href="https://uzh.ch">Universit&auml;t Z&uuml;rich</a></div><div class="event_type"><a class="external" href="https://www.math.uzh.ch/aa/?events">Neuchatel - St.Gallen - Zurich Seminar in Coding Theory and Cryptography</a></div><br /><div class="event_speaker">Yves Krähenbühl (University of Zurich)</div><div class="event_title">Overview of Classes of MDP Convolutional Codes and Their Erasure Decoding Algorithms <span class="abstract-button do_not_print">abstract</span></div><div class="abstract-text" style="display: none"><strong>Abstract:</strong><br />Convolutional codes are a class of codes particularly well-suited for data transmission over erasure channels, commonly used in multimedia traffic over the internet. This thesis compares three erasure decoding algorithms for convolutional codes: The forward and backward decoding algorithm, the low delay decoding algorithm for modules, and the low delay decoding algorithm for linear systems. Our primary focus is on comparing these algorithms in terms of their delays, computational complexities, and erasure recovery capabilities. Additionally, we discuss various classes of MDP convolutional codes, highlighting their advantageous properties in achieving an optimal performance with the three erasure decoding algorithms. Through two simulations, we assess the performance of the decoding algorithms over different erasure channel models simulating practical scenarios. In a first simulation, we evaluate the recovery capability of the algorithms using their respective optimal convolutional code classes and compare them with MDS linear block codes of the same rate. The results show that convolutional codes consistently match or even outperform MDS block codes over the different channel models. In a second simulation, we use longer received codewords relative to the largest sliding window of a convolutional code, revealing that the recovery process is influenced not only by the quantity of erasures but also by their distribution. Notably, early erasures have a minimal influence on the remainder of the decoding process. This thesis provides valuable insights into the selection and optimization of erasure decoding algorithms for convolutional codes, offering practical implications for multimedia data transmission over the internet.</div><div class="event_time">16:15 &bull; UZH Irchel, Winterthurerstrasse 190, Z&uuml;rich, Building Y27, Room H 25</div>        </div>        <div class="day_content_item normal"><div class="event_location uni_zh"><a class="external" href="https://ethz.ch">ETH Z&uuml;rich</a></div><div class="event_type"><a class="external" href="https://math.ethz.ch/sam/news-and-events/zhacm-colloquia">Zurich Colloquium in Applied and Computational Mathematics</a></div><br /><div class="event_speaker">Prof. Dr. H. Gimperlein (Leopold-Franzens-Universität Innsbruck)</div><div class="event_title">Boundary integral equations in space and time: Higher order Galerkin methods and applications <span class="abstract-button do_not_print">abstract</span></div><div class="abstract-text" style="display: none"><strong>Abstract:</strong><br />Boundary integral formulations are well-known to lead to efficient numerical methods for time-independent scattering and emission problems.  In this talk we consider corresponding formulations for the time-dependent acoustic and elastic wave equations. We survey recent work on space-time Galerkin methods for the numerical solution, including higher order approximations by h- and hp-versions, a posteriori error estimates and adaptive mesh refinements, and illustrate them for applications in traffic noise.</div><div class="event_time">16:30 &bull; ETH Zentrum, R&auml;mistrasse 101, Z&uuml;rich, Building HG, Room E 1.2</div>        </div>        <div class="day_content_item normal"><div class="event_location uni_zh"><a class="external" href="https://uzh.ch">Universit&auml;t Z&uuml;rich</a></div><div class="event_type"><a class="external" href="https://math.ethz.ch/s/seminar-on-stochastic-processes">Seminar on Stochastic Processes</a></div><br /><div class="event_speaker">Prof. Dr. Grégory Miermont (ENS Lyon)</div><div class="event_title">Compact Brownian surfaces <span class="abstract-button do_not_print">abstract</span></div><div class="abstract-text" style="display: none"><strong>Abstract:</strong><br />We describe the compact scaling limits of uniformly random quadrangulations with boundaries on a surface of arbitrary fixed genus. These limits, called Brownian surfaces, are homeomorphic to the surface of the given genus with or without boundaries depending on the scaling regime of the boundary perimeters of the quadrangulation. They are constructed by appropriate gluings of pieces derived from Brownian geometrical objects (the Brownian plane and half-plane). In this talk, I will review their definition and discuss possible alternative constructions. This is based on joint work with Jérémie Bettinelli. </div><div class="event_time">17:15 &bull; UZH Irchel, Winterthurerstrasse 190, Z&uuml;rich, Building Y27, Room  H12</div>        </div>      </div>    </div>    <div class="day">      <div class="day_title">        Thursday, September 28      </div>      <div class="day_content">        <div class="day_content_item normal"><div class="event_location uni_zh"><a class="external" href="https://ethz.ch">ETH Z&uuml;rich</a></div><div class="event_type"><a class="external" href="https://math.ethz.ch/fim/activities/nachdiplom-lectures">Nachdiplomvorlesung</a></div><br /><div class="event_speaker"> Javier Fresán (École Polytechnique, Palaiseau)</div><div class="event_title">E-functions and geometry</div><div class="event_time">10:15 &bull; ETH Zentrum, R&auml;mistrasse 101, Z&uuml;rich, Building HG, Room G 43</div>        </div>        <div class="day_content_item normal"><div class="event_location uni_ba"><a class="external" href="https://unibas.ch">Universität Basel</a></div><div class="event_type"><a class="external" href="https://dmi.unibas.ch/de/forschung/mathematik/zahlentheorie/">Seminar Zahlentheorie</a></div><br /><div class="event_speaker"> Gabriele  Bogo (TU Darmstadt)</div><div class="event_title">Supersingular abelian surfaces and orthogonal polynomials <span class="abstract-button do_not_print">abstract</span></div><div class="abstract-text" style="display: none"><strong>Abstract:</strong><br />An abelian surface in characteristic p>0 is called supersingular if it is isogenous to the product of two elliptic curves with trivial p-torsion. I will show that the supersingular locus of the mod p reduction of genus zero curves in Hilbert modular surfaces can be described in terms of zeros of orthogonal polynomials. The proof is based on the theory of twisted modular forms. This is joint work with Yingkun Li.</div><div class="event_time">14:15 &bull; Universität Basel, Spiegelgasse 1, SR 00.003</div>        </div>        <div class="day_content_item normal"><div class="event_location uni_zh"><a class="external" href="https://ethz.ch">ETH Z&uuml;rich</a></div><div class="event_type"><a class="external" href="https://www.math.uzh.ch/mat675.1">PDE and Mathematical Physics</a></div><br /><div class="event_speaker">Jaemin Park (University of Basel)</div><div class="event_title">Small scale creation for the 2D Boussinesq Equations <span class="abstract-button do_not_print">abstract</span></div><div class="abstract-text" style="display: none"><strong>Abstract:</strong><br />In this presentation, we will examine the long-term behaviors of the two- imensional incompressible Boussinesq equations without thermal diffusion. While the Boussinesq equations exhibit energy conservation, the flow\'s small-scale creation may induce the growth of finer norms in the solutions over time. During this talk, we will construct smooth initial data to demonstrate such norm growth phenomena, both with and without considering kinematic viscosity. Additionally, we will extend our results to provide an example of temperature patch solutions, the curvature and perimeter of which increase as time progresses. This work is a joint work with A. Kiselev and Y. Yao.</div><div class="event_time">16:15 &bull; ETH Zentrum, R&auml;mistrasse 101, Z&uuml;rich, Building HG, Room G 19.2</div>        </div>      </div>    </div>    <div class="day">      <div class="day_title">        Friday, September 29      </div>      <div class="day_content">        <div class="day_content_item normal"><div class="event_location uni_ba"><a class="external" href="https://unibas.ch">Universität Basel</a></div><div class="event_type"><a class="external" href="https://dmi.unibas.ch/de/forschung/mathematik/seminar-in-numerical-analysis">Seminar in Numerical Analysis</a></div><br /><div class="event_speaker"> Rüdiger Kempf (U Bayreuth)</div><div class="event_title">A Kernel-based Multilevel Approximation Method for Higher-Dimensional Problems on Sparse Grids <span class="abstract-button do_not_print">abstract</span></div><div class="abstract-text" style="display: none"><strong>Abstract:</strong><br />Reproducing kernel Hilbert spaces (RKHSs) and the closely related kernel methods are well-established and well-studied tools in classical approximation theory. More recently, they see many uses in other problems in applied and numerical analysis.In machine learning, support vector machines heavily rely on RKHSs. For neural networks Barron spaces are connected to certain RKHSs and offer a possibility for a theoretical analysis of these networks.Another application of RKHSs is in high(er)-dimensional approximation. For instance in the field of quasi Monte-Carlo methods, kernel-techniques are used to derive an error analysis for high-dimensional quadrature rules. We also developed a novel kernel-based approximation method for higher-dimensional meshfree function reconstruction, based on Smolyak operators.In this talk I will provide an introduction into the theory of RKHSs, their kernels and associated kernel methods. In particular, I will focus on a multiscale approximation scheme for rescaled radial basis functions. This method will then be used to derive the new tensor product multilevel method for higher- dimensional meshfree approximation, which I will discuss in detail.</div><div class="event_time">11:00 &bull; Universität Basel, Spiegelgasse 5, Seminarraum 05.001</div>        </div>        <div class="day_content_item normal"><div class="event_location uni_zh"><a class="external" href="https://ethz.ch">ETH Z&uuml;rich</a></div><div class="event_type"><a class="external" href="https://math.ethz.ch/s/number-theory-seminar">Number Theory Seminar</a></div><br /><div class="event_speaker"><a class="external" href="https://math.ethz.ch/people/vkuperber">Dr. Vivian Kuperberg (ETH Z&uuml;rich, Switzerland)</a></div><div class="event_title">Consecutive sums of two squares in arithmetic progressions <span class="abstract-button do_not_print">abstract</span></div><div class="abstract-text" style="display: none"><strong>Abstract:</strong><br />In 2000, Shiu proved that there are infinitely many primeswhose last digit is 1 such that the next prime also ends in a 1.However, it is an open problem to show that there are infinitely manyprimes ending in 1 such that the next prime ends in 3.In this talk, we\'ll instead consider the sequence of sums of twosquares in increasing order. In particular, we\'ll show that there areinfinitely many sums of two squares ending in 1 such that the nextsum of two squares ends in 3. We\'ll show further that all patterns oflength 3 occur infinitely often: for any modulus q, every sequence (amod q, b mod q, c mod q) appears infinitely often among consecutivesums of two squares. We\'ll discuss some of the proof techniques, andexplain why they fail for primes. Joint work with Noam Kimmel.</div><div class="event_time">14:15 &bull; ETH Zentrum, R&auml;mistrasse 101, Z&uuml;rich, Building HG, Room G 43</div>        </div>        <div class="day_content_item normal"><div class="event_location uni_zh"><a class="external" href="https://ethz.ch">ETH Z&uuml;rich</a></div><div class="event_type"><a class="external" href="https://stat.ethz.ch/events/research_seminar">Research Seminar in Statistics</a></div><br /><div class="event_speaker"> Leonardo Egidi  (University of Trieste)</div><div class="event_title">Prediction, skepticism, and the Bayes Factory <span class="abstract-button do_not_print">abstract</span></div><div class="abstract-text" style="display: none"><strong>Abstract:</strong><br />Nowadays a Bayesian model needs to be reproducible, generative, predictive, robust, computationallyscalable, and able to provide sound inferential conclusions.  In this wide framework, Bayes factors still represent one of the most well-known and commonly adopted tools to perform model selection and hypothesis testing; however, they are usually criticized due to their intrinsic lack of calibration, and they are rarely used to measure the predictive accuracy arising from competing models. We propose two distinct approaches relying on BFs from our most recent research. With regard to prediction, we propose a new algorithmic protocol to transform Bayes factors into measures that evaluate the pure and intrinsic predictive capabilities of models in terms of posterior predictive distributions, by assessing some preliminary theoretical properties (joint work with Ioannis Ntzoufras).Then, regarding the analysis of replication studies (Held, 2020), we follow the stream outlined by Pawel and Held (2022) and propose a skeptical mixture prior which represents the prior of an investigator who is unconvinced by the original findings. Its novelty lies in the fact that it incorporates skepticism while controlling for prior-data conflict (Egidi et al., 2021). Consistency properties of the resulting skeptical BF are provided together with a thorough analysis of the main features of our proposal (joint work with Guido Consonni).Short Bibliography:Egidi, L., Pauli, F., & Torelli, N. (2022). Avoiding prior–data conflict in regression models via mixture priors. Canadian Journal of Statistics, 50(2), 491-510.Held, L. (2020). A new standard for the analysis and design of replication studies. Journal of the Royal Statistical Society Series A: Statistics in Society, 183(2), 431-448.Pawel, S., & Held, L. (2022). The sceptical Bayes factor for the assessment of replication success. Journal of the Royal Statistical Society Series B: Statistical Methodology, 84(3), 879-911.</div><div class="event_time">15:15 &bull; ETH Zentrum, R&auml;mistrasse 101, Z&uuml;rich, Building HG, Room G 19.1</div>        </div>      </div>    </div>    <div class="day">      <div class="day_title">        Saturday, September 30      </div>      <div class="day_content">        <div class="day_content_item normal"><div class="event_location uni_zh"><a class="external" href="https://ethz.ch">ETH Z&uuml;rich</a></div><div class="event_type"><a class="external" href="https://memento.epfl.ch/maths/">External FLAIR Seminar </a></div><br /><div class="event_speaker"> Spencer Frei (UC Berkeley)</div><div class="event_title">Implicit bias and benign overfitting for neural networks in high dimensions <span class="abstract-button do_not_print">abstract</span></div><div class="abstract-text" style="display: none"><strong>Abstract:</strong><br />Benign overfitting, the phenomenon where interpolating models generalize well in the presence of noisy data, was first observed in neural networks trained by gradient descent.  In this talk we go over some recent work towards understanding this surprising phenomenon.   We first describe an implicit regularization effect of gradient descent in two-layer neural networks when trained on high-dimensional datasets.  We show that in this setting, gradient descent finds solutions which have small rank, despite the lack of explicit regularization to encourage such structure.  We then consider the generalization error of trained two-layer networks when the data comes from a high-dimensional mixture model where a constant fraction of the training labels are uniformly random labels.  In this setting, we show that neural networks indeed exhibit benign overfitting: they can be driven to zero training error, perfectly fitting the noisy training labels, and simultaneously achieve minimax-optimal test error.  In contrast to previous work on benign overfitting that require linear or kernel-based predictors, our analysis holds in a setting where both the model and learning dynamics are fundamentally nonlinear.  Based on previous and upcoming work with Peter Bartlett, Niladri Chatterji, Wei Hu, Nati Srebro, and Gal Vardi.</div><div class="event_time">13:15 &bull; EPF Lausanne GA 3 21</div>        </div>      </div>    </div>  </div>  <div id="sms_footer">    address: Alessandra Naldi Windler, Departement Mathematik, HG G 51.4, ETH Zurich, Raemistrasse 101, CH-8092 Zurich<br />    email: <a href="mailto:bulletin@math.ch">bulletin@math.ch</a> | phone: +41 44 632 3684 | url: <a href="https://math.ch">https://math.ch</a>    <br /><br /><img src="//www2.math.ethz.ch/smsjs/sms_logo.gif" width="150" alt="SMS/SMG Logo" title="SMS/SMG Logo" />  </div></div>';