# Greta Panova

Professor

USC

Organized by the CRM in collaboration with the Institut des sciences mathématiques (ISM), the Colloque des sciences mathématiques du Québec offers a forum for mathematicians of great reputation, who are invited to give lectures of current and general interest, and accessible to the entire Quebec mathematical community. The tradition is that these lectures are as qualitative and non-technical as possible in order to be accessible to general graduate students in mathematics and statistics.

Zoom link

December 8, 2023 from 15:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **Greta Panova (USC)**

Algebraic Combinatorics studies objects and quantities originating in Algebra, Representation Theory and Algebraic Geometry via combinatorial methods, finding formulas and neat interpretations. Some of its feats include the hook-length formula for the dimension of an irreducible symmetric group (S_n) module, or the Littlewood-Richardson rule to determine multiplicities of GL irreducibles in tensor products. Yet some natural multiplicities elude us, among them the Kronecker coefficients for the decomposition of tensor products of S_n irreducibles, and the plethysm coefficients for compositions of GL modules. Answering those questions could help Geometric Complexity Theory establish lower bounds for the far reaching goal to show that P \neq NP.

We will discuss how Computational Complexity Theory provides a theoretical framework for understanding what kind of formulas or rules we could actually have. We will use this to show that the square of a symmetric group character could not have a combinatorial interpretation.

Based on joint works with Christian Ikenmeyer and Igor Pak.

**Address**

Change display

Start time | Title | Speaker |
---|---|---|

2023-12-08 15:30 | Computational Complexity in Algebraic Combinatorics | Greta Panova (USC) |

December 8, 2023 from 15:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **Greta Panova (USC)**

Algebraic Combinatorics studies objects and quantities originating in Algebra, Representation Theory and Algebraic Geometry via combinatorial methods, finding formulas and neat interpretations. Some of its feats include the hook-length formula for the dimension of an irreducible symmetric group (S_n) module, or the Littlewood-Richardson rule to determine multiplicities of GL irreducibles in tensor products. Yet some natural multiplicities elude us, among them the Kronecker coefficients for the decomposition of tensor products of S_n irreducibles, and the plethysm coefficients for compositions of GL modules. Answering those questions could help Geometric Complexity Theory establish lower bounds for the far reaching goal to show that P \neq NP.

We will discuss how Computational Complexity Theory provides a theoretical framework for understanding what kind of formulas or rules we could actually have. We will use this to show that the square of a symmetric group character could not have a combinatorial interpretation.

Based on joint works with Christian Ikenmeyer and Igor Pak.

**Address**

Change display

November 24, 2023 from 15:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **Leilei Zeng (University of Waterloo)**

Observational cohort studies of chronic disease involve the recruitment and follow-up of a sample of individuals with the goal of learning about the course of the disease, the effect of fixed and time-varying risk factors. Analysis of this information is often facilitated by using multistate models with intensity functions governing transition between disease states. Chronic disease studies often involve conditions for recruitment, for example incident cohort involves individuals who are healthy at accrual, prevalent cohort samples individuals who have already developed the disease, and a length biased sampling includes individual who are alive at the time of recruitment. In this talk we discuss the impact of ignoring state-dependent sampling in life history analysis and the ways of addressing the issue using auxiliary information. A longitudinal study of aging and cognition among religious sisters is used to illustrate the related methodology.

**Address**

November 17, 2023 from 15:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **Christophe Breuil (Université Paris-Saclay)**

Let n be an integer greater than or equal to 2. The Langlands program for GLn connects n-dimensional representations of Galois groups and infinite dimensional representations of GLn. For the purposes of number theory, we are led to consider all these representations on vector spaces over fields of characteristic p for p an arbitrary prime number. On the GLn side, this leads in particular to the following problem: understand and construct the - or some - representations of GLn(K) in characteristic p where K is a finite extension of the field of p-adic numbers Qp (for the same prime number p!), and most importantly those of these representations which appear on cohomology spaces. This problem has challenged experts for more than 20 years and is still largely open even for GL2(K). I will recall the history, the difficulties encountered, and will state some recent results for GL2(K)

**Address**

November 10, 2023 from 15:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **James V. Burke (University of Washington)**

Optimization problems have an enormous variety and complexity and solving them requires techniques for exploiting their underlying mathematical structure. The modeler needs to balance model complexity with computational tractability as well as viable techniques for post optimal analysis and stability measures.

In this talk we describe the convex-composite modeling framework which covers a broad range of optimization problems including nonlinear programming, feasibility, minimax optimization, sparcity optimization, feature selection, Kalman smoothing, parameter selection, and nonlinear maximum likelihood to name a few. The goal is to identify and exploit the underlying convexity that a given problem may possess since convexity allows one to tap into the very rich theoretical foundation as well as the wide range of highly efficient numerical methods available for convex problems. The systematic study of convex-composite problems began in the 1970's concurrent with the emergence of modern nonsmooth variational analysis. The synergy between these ideas was natural since convex-composite functions are neither convex nor smooth. The recent resurgence in interest for this problem class is due to emerging methods for approximation, regularization and smoothing as well as the relevance to a number of problems in global health, environmental modeling, image segmentation, dynamical systems, signal processing, machine learning, and AI. In this talk we review the convex-composite problem structure and variational properties. We then discuss algorithm design and if time permits, we discuss applications to filtering methods for signal processing.

**Address**

November 3, 2023 from 15:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **Sam Payne (University of Texas at Austin)**

Algebraic geometry studies solution sets of polynomial equations. For instance, over the complex numbers, one may examine the topology of the solution set, whereas over a finite field, one may count its points. For polynomials with integer coefficients, these two fundamental invariants are intimately related via cohomological comparison theorems and trace formulas for the action of Frobenius. I will present recent results regarding point counts over finite fields and the cohomology of moduli spaces of curves that resolve longstanding questions in algebraic geometry and confirm more recent predictions from the Langlands program.

**Address**

October 27, 2023 from 15:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **Stéphane Jaffard (Université Paris-Est Crétell)**

The detection of specific oscillatory behaviours in signals is a key issue in turbulence, gravitational waves, or physiological data, where they turn out to be the signature of important phenomena localized in time or space. We will show how some recently introduced methods of harmonic analysis allow us to characterize and classify such behaviors, and ultimately yield numerical methods to perform their detection.

**Address**

October 20, 2023 from 15:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **Sergei Tabachnikov (Pennsylvania State University)**

Optical properties of conics have been known since the classical antiquity. The reflection in an ideal mirror is also known as the billiard reflection and, in modern terms, the billiard inside an ellipse is completely integrable. The interior of an ellipse is foliated by confocal ellipses that are its caustics: a ray of light tangent to a caustic remains tangent to it after reflection (“caustic” means burning).

I shall explain these classic results and some of their geometric consequences, including the Ivory lemma asserting that the diagonals of a curvilinear quadrilateral made by arcs of confocal ellipses and hyperbolas are equal (this lemma is in the heart of Ivory's calculation of the gravitational potential of a homogeneous ellipsoid). Other applications include the Poncelet Porism, a famous theorem of projective geometry that has celebrated its bicentennial, and its lesser known ramifications, such as the Poncelet Grid theorem and the related circle patterns and configuration theorems.

**Address**

October 13, 2023 from 15:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **Lia Bronsard (McMaster University)**

We study the Nakazawa-Ohta ternary inhibitory system, which describes domain morphologies in a triblock copolymer as a nonlocal isoperimetric problem for three interacting phase domains. The free energy consists of two parts: the local interface energy measures the total perimeter of the phase boundaries, while a longer-range Coulomb interaction energy reflects the connectivity of the polymer chains and promotes splitting into micro-domains. We consider global minimizers on the two-dimensional torus, in a limit in which two of the species have vanishingly small mass but the interaction strength is correspondingly large. In this limit there is splitting of the masses, and each vanishing component rescales to a minimizer of an isoperimetric problem for clusters in 2D. Depending on the relative strengths of the coefficients of the interaction terms we may see different structures for the global minimizers, ranging from a lattice of isolated simple droplets of each minority species to double-bubbles or core-shells. These results have led to a new type of partitioning problem that I will also introduce. These represent work with S. Alama, with X. Lu, and C. Wang, as well as with S. Vriend.

**Address**

October 6, 2023 from 15:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **Zhou Zhou (University of Toronto)**

Understanding the time-varying structure of complex temporal systems is one of the main challenges of modern time series analysis. In this talk, I will demonstrate that a wide range of short-range dependent non-stationary and nonlinear time series can be well approximated globally by a white-noise-driven auto-regressive (AR) process of slowly diverging order. Uniform statistical inference of the latter AR structure will be discussed through a class of high-dimensional L2 tests. I will further discuss applications of the AR approximation theory to globally optimal short-term forecasting, efficient estimation, and resampling inference under complex temporal dynamics.

**Address**

September 29, 2023 from 15:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **Nicolai Krylov (University of Minnesota)**

We prove the existence of strong solutions of It\^o's stochastic time dependent equations with irregular diffusion and drift terms of Morrey spaces. Strong uniqueness is also discussed. The results are new even if there is no drift. The results are based on the solvability of parabolic equations with Morrey

drift in Morrey spaces, which is also new.

**Address**

September 15, 2023 from 15:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **Maksym Radziwill (Northwestern)**

The limiting distributions for maxima of independent random variables have been classified during the first half of last century. This classification does not extend to strong interactions, in particular to the flurry of processes with

natural logarithmic (or multiscale) correlations. These include branching random walks or the 2d Gaussian free field. More recently, Fyodorov, Hiary and Keating (2012) exhibited new examples of log-correlated phenomena in number theory and random matrix theory. As a result (and as a testing ground of their observations) they have formulated very precise conjectures about maxima of the characteristic polynomial of random matrices, and the maximum of L-functions on typical interval the critical line. I will describe the recent progress towards these conjectures in both the random and deterministic setting.

**Address**

May 19, 2023 from 15:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **Alex Iosevich (University of Rochester)**

The basic question we ask is, how large does a subset of a given vector space need to be to ensure that it contains a congruent copy of a given point configuration. In Euclidean space, the size is measured in terms of Hausdorff dimension. In finite fields, the counting measure is used. We are going to survey a variety of recent and not so recent results, as well as connections between them. Emerging connections with learning theory will be mentioned as well.

**Address**

May 12, 2023 from 15:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **Erica E.M. Moodie (McGill University)**

Estimating individualized treatment rules is challenging, as the treatment effect heterogeneity of interest often suffers from low power. This motivates the use of very large datasets such as those from multiple health systems or multicentre studies, which may raise concerns of data privacy. In this talk, I will introduce a statistical framework for of estimation individualized treatment rules and show how distributed regression can be used in combination with dynamic weighted regression to find an optimal individualized treatment rule whilst obscuring individual-level data. The robustness of this approach and its flexibility to address local treatment practices will be shown in simulation. The work is motivated by, and illustrated with, an analysis of the U.K.’s Clinical Practice Research Datalink on the treatment of depression.

**Address**

May 5, 2023 from 16:00 to 17:00 (Montreal/EST time) On location

Colloquium presented by **Mikhail Karpukhin (University College London)**

Eigenvalues of the Laplace operator of Euclidean domains govern many physical phenomena, including heat flow and sound propagation. In particular, various inequalities for Laplace eigenvalues have fascinated mathematicians since XIXth century. The following question was first formulated by Lord Rayleigh in his “Theory of sound”: which planar domain of given area has the lowest first Dirichlet eigenvalue? This is an example of an isoperimetric eigenvalue problem for planar domains. The focus of the present talk is on more general isoperimetric problems, where one considers surfaces equipped with Riemannian metrics. More specifically, sharp upper bounds for Laplace and Steklov eigenvalues have been an active area of research for the past decade, largely due to their fascinating connection to fundamental geometric objects, minimal surfaces. We will survey recent results exploring the applications of this connection both to minimal surface theory and to isoperimetric eigenvalue problems, culminating in a surprising link between Laplace and Steklov spectra.

**Address**

April 28, 2023 from 15:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **José A. Carrillo (University of Oxford)**

I will present a survey of micro, meso and macroscopic models where repulsion and attraction effects are included through pairwise potentials. I will discuss their interesting mathematical features and applications in mathematical biology and engineering. Qualitative properties of local minimizers of the interaction energies are crucial in order to understand these complex behaviors. I will showcase the breadth of possible applications with three different phenomena in applications: segregation, phase transitions and consensus.

**Address**

April 21, 2023 from 15:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **David London (Université de Montréal)**

The standard model (SM) of particle physics explains nearly all experimental results to date. There is no doubt that it is correct. However, for a variety of reasons it is understood to be incomplete – there must be physics beyond the SM. In this talk, I provide a brief review of the SM, discuss the reasons we believe it is incomplete, and present some examples of my contributions over the years to this search for new physics.

**Address**

March 31, 2023 from 15:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **Michael Hrusak (Universidad Nacional Autónoma de México-Morelia)**

We shall discuss some recent applications of set-theoretic and model-theoretic methods to the study of topological groups. In particular, we shall outline how ultra powers can be used to solve old problems of Comfort and van Douwen and introduce a new set-theoretic axiom to study convergence properties in topological groups. If time permits we may also briefly mention the use of Fraissé theory in the study of groups of homeomorphisms.

**Address**

March 24, 2023 from 15:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **Eugenia Malinnikova (Stanford University)**

In this talk we will give an overview of some recent results on unique continuation property at infinity for solutions of elliptic equations. Our first result is an unexpected uniqueness property for discrete harmonic functions. This property is connected to Anderson localization for Anderson-Bernoulli model in dimensions two and three. We will explain this connection. Another result is the solution of the Landis conjecture on the decay of the real-valued solutions of the Schrodinger equation with bounded potential. The talk is based on joint works with Buhovsky, Logunov, Sodin, Nadirashvili, and Nazarov.

**Address**

March 17, 2023 from 15:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **Matilde Lalin (Université de Montréal)**

The divisor function gives the number of positive divisors of a natural number. How can we go about understanding the behavior of this function when going over the natural numbers? In this talk we will discuss strategies to better understand this function, issues related to the distribution of these values, and connections to the Riemann zeta function and some groups of random matrices.

**Address**

March 10, 2023 from 15:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **Vasilisa Shramchenko (Sherbrooke University)**

Riemann surfaces and algebraic curves are ubiquitous in mathematics. Without attempting a review of the variety of ways in which they are useful, we will look at some specific examples and discuss some curious links with physics and mathematical physics, including a link between the quantum harmonic oscillator and moduli spaces of Riemann surfaces, and a link between the periodic Toda chain, Chebyshev polynomials, and Painlevé equations.

**Address**

February 24, 2023 from 15:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **Michael Groechenig (University of Toronto)**

A representation of a group G is said to be rigid, if it cannot be continuously deformed to a non-isomorphic representation. If G happens to be the fundamental group of a complex projective manifold, rigid representations are conjectured (by Carlos Simpson) to be of geometric origin. In this talk I will outline the basic properties of rigid local systems and discuss several consequences of Simpson‘s conjecture. I will then outline recent progress on these questions (joint work with Hélène Esnault) and briefly mention applications to geometry and number theory such as the recent resolution of the André-Oort conjecture by Pila-Shankar-Tsimerman.

**Address**

February 10, 2023 from 15:30 to 18:30 (Montreal/EST time) On location

Colloquium presented by **Simone Brugiapaglia (Concordia University)**

Deep learning is having a profound impact on industry and scientific research. Yet, while this paradigm continues to show impressive performance in a wide variety of applications, its mathematical foundations are far from being well established. In this talk, I will present recent developments in this area by illustrating two case studies.

First, motivated by applications in cognitive science, I will present “rating impossibility" theorems. They identify frameworks where deep learning is provably unable to generalize outside the training set for the seemingly simple task of learning identity effects, i.e. classifying whether pairs of objects are identical or not.

Second, motivated by applications in scientific computing, I will illustrate “practical existence" theorems. They combine universal approximation results for deep neural networks with compressed sensing and high-dimensional polynomial approximation theory. As a result, they yield sufficient conditions on the network architecture, the training strategy, and the number of samples able to guarantee accurate approximation of smooth functions of many variables.

Time permitting, I will also discuss work in progress and open questions.

**Address**

January 27, 2023 from 15:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **Arul Shankar (University of Toronto)**

The Goldfeld and Katz--Sarnak conjectures predict that 50% of elliptic curves have rank 0, that 50% have rank 1, and that the average rank of elliptic curves is 1/2 (the remaining 0% of elliptic curves not interfering in the average). Successive works of Brumer, Heath-Brown, and Young, have approached this problem by studying the central values of the L functions of elliptic curves. In this talk, we will take an algebraic approach, in which we study the ranks of elliptic curves via studying their Selmer groups.

Poonen and Stoll developed a beautiful model for the behaviours of $p$-Selmer groups of elliptic curves, and gave heuristics for all moments of the sizes of these groups.

In this talk, I will describe joint work with Manjul Bhargava and Ashvin Swaminathan, in which we prove that the second moment of the size of the 2-Selmer groups of elliptic curves is bounded above by 15 (which is the constant predicted by Poonen--Stoll).

There won’t be a Zoom broadcast for this conference

**Address**

January 20, 2023 from 15:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **Jaroslav Nešetřil (Charles University, Prague)**

Several combinatorial problems are treated in the context of model theory.

We survey three such instances which were investigated recently, coming

from Ramsey theory, sparsity of graphs and limits of sequences of structures.

These are diverse areas but share some properties where the connection to

model theory is non-trivial and interesting. It also presents several open

problems of interest to both combinatorics and model theory.

**Address**

January 13, 2023 from 15:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **Joshua Zahl**

A Kakeya set is a compact subset of R^n that contains a unit line segment pointing in every direction. The Kakeya conjecture asserts that such sets must have dimension n. This conjecture is closely related to several open problems in harmonic analysis, and it sits at the base of a hierarchy of increasingly difficult questions about the behavior of the Fourier transform in Euclidean space.

There is a special class of Kakeya sets, called sticky Kakeya sets. Sticky Kakeya sets exhibit an approximate self-similarity at many scales, and sets of this type played an important role in Katz, Łaba, and Tao's groundbreaking 1999 work on the Kakeya problem. In this talk, I will discuss a special case of the Kakeya conjecture, which asserts that sticky Kakeya sets must have dimension n. I will discuss the proof of this conjecture in dimension 3. This is joint work with Hong Wang.

**Address**

December 9, 2022 from 15:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **Yuri Bakhtin (New York University)**

I am interested in stationary distributions for the Burgers equation with random forcing. I will first consider an oversimplified random dynamical system to illustrate the power of a general approach based on the so-called pullback procedure. For the Burgers equation, which is a basic evolutionary stochastic PDE of Hamilton-Jacobi type related to fluid dynamics, growth models, and the KPZ equation, one can realize this approach via studying long-term properties of random Lagrangian action minimizers and directed polymer measures in random environments. The compact space case was studied in 2000's. This talk is based on my work on the noncompact case, joint with Eric Cator, Kostya Khanin, Liying Li.

**Address**

December 2, 2022 from 15:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **Amnon Besser (Ben Gurion University)**

The problem of finding rational or integer solutions to polynomial equations is one of the oldest problems in mathematics and is one of the key driving forces in the development of Number Theory. In the last 15 years new methods were developed that can sometimes effectively solve this problem. These methods attempt to find the solutions inside the larger set of solutions of the same equation in the field of p-adic numbers as the vanishing set of some computable function. When these methods work they give the rational solutions to arbitrarily large p-adic precision, which usually suffices to rigorously recover the full set of solutions.

I will survey the new methods, originating from the work of Kim and from the more recent work of Lawrence and Venkatesh. I will then explain my work with Muller and Srinivasan that uses a p-adic version of the notion of norms on line bundles and associated heights, as used for example in arithmetic dynamics, to give a new approach to some Kim type results.

The conference will not be broadcasted via Zoom.

**Address**

November 25, 2022 from 15:30 to 16:30 (Montreal/EST time) Zoom meeting

Colloquium presented by **Rebecca Steorts (Duke University)**

Whether the goal is to estimate the number of people that live in a congressional district, to estimate the number of individuals that have died in an armed conflict, or to disambiguate individual authors using bibliographic data, all these applications have a common theme — integrating information from multiple sources. Before such questions can be answered, databases must be cleaned and integrated in a systematic and accurate way, commonly known as record linkage, de-duplication, or entity resolution. In this article, we review motivational applications and seminal papers that have led to the growth of this area. Specifically, we review the foundational work that began in the 1940’s and 50’s that have led to modern probabilistic record linkage. We review clustering approaches to entity resolution, semi- and fully supervised methods, and canonicalization, which are being used throughout industry and academia in applications such as human rights, official statistics, medicine, citation networks, among others. Finally, we discuss current research topics of practical importance. This is joint work with Olivier Binette.

**Address**

November 18, 2022 from 15:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **Robin Neumayer (Carnegie Mellon University)**

Among all subsets of Euclidean space with a fixed volume, balls have the smallest perimeter. Furthermore, any set with nearly minimal perimeter is geometrically close, in a quantitative sense, to a ball. This latter statement reflects the quantitative stability of balls with respect to the perimeter functional. We will discuss recent advances in quantitative stability and applications in various contexts. The talk includes joint work with several collaborators and will be accessible to a broad research audience.

**Address**

November 11, 2022 from 15:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **Jan Vonk (Leiden University)**

This is the story of an old equation of Diophantus, which will take us on an excursion along a branch of number theory that stretches over a large part of the subject's history. It will be our motivation for a friendly introduction to some modern developments on torsion and ranks of elliptic curves.

**Address**

November 4, 2022 from 15:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **Jacob Bernstein (Johns Hopkins University)**

Given a submanifold of Euclidean space, Colding and Minicozzi defined its entropy to be the supremum of the Gaussian weighted surface areas of all of its translations and dilations. While initially introduced to study singularities of mean curvature flow, it has proven to be an interesting geometric measure of complexity. In this talk I will survey some of the recent progress made on studying the Colding-Minicozzi entropy of hypersurfaces. In particular, I will discuss a series of work by Lu Wang and myself showing closed hypersurfaces with small entropy are simple in various senses.

**Address**

October 28, 2022 from 15:30 to 16:30 (Montreal/EST time)

Colloquium presented by **Derek Bingham (Simon Fraser University)**

Computer models are often used to explore physical systems. Increasingly, there are cases where the model is fast, the code is not readily accessible to scientists, but a large suite of model evaluations is available. In these cases, an “emulator” is used to stand in for the computer model. This work was motivated by a simulator for the chirp mass of binary black hole mergers where no output is observed for large portions of the input space and more than 10^6 simulator evaluations are available. This poses two problems: (i) the need to address the discontinuity when observing no chirp mass; and (ii) performing statistical inference with a large number of simulator evaluations. The traditional approach for emulation is to use a stationary Gaussian process (GP) because it provides a foundation for uncertainty quantification for deterministic systems. We explore the impact of the choices when setting up the deep GP on posterior inference, apply the proposed approach to the real application and propose a sequential design approach for identifying new simulations.

**Address**

October 14, 2022 from 15:30 to 16:30 (Montreal/EST time)

Colloquium presented by **Yevgeny Liokumovich (University of Toronto, Canada)**

I will describe two new isoperimetric inequalities for k-dimensional submanifolds of R^n or a Banach space. As a consequence of one we obtain a new systolic inequality that was conjectured by Larry Guth. As a consequence of another, we obtain an asymptotic formula for volumes of minimal submanifolds that was conjectured by Mikhail Gromov. The talk is based on joint works with Boris Lishak, Alexander Nabutovsky and Regina Rotman; Fernando Marques and Andre Neves; Larry Guth.

**Address**

October 7, 2022 from 15:30 to 16:30 (Montreal/EST time)

Colloquium presented by **Balint Virag (University of Toronto)**

Consider Z^2, and assign a random length of 1 or 2 to each edge based on independent fair coin tosses. The resulting random geometry, first passage percloation, is conjectured to have a scaling limit.

Most random plane geometric models (including hidden geometries) should have the same scaling limit.

I will explain the basics of the limiting geometry, the "directed landscape", the central object in the class of models named after

Kardar, Parisi and Zhang.

**Address**

September 30, 2022 from 15:30 to 16:30 (Montreal/EST time) Zoom meeting

Colloquium presented by **Pengfei Li (University of Waterloo)**

Capture-recapture experiments are widely used to collect data needed to estimate the abundance of a closed population. To account for heterogeneity in the capture probabilities, Huggins (1989) and Alho (1990) proposed a semiparametric model in which the capture probabilities are modelled parametrically and the distribution of individual characteristics is left unspecified. A conditional likelihood method was then proposed to obtain point estimates and Wald-type confidence intervals for the abundance. Empirical studies show that the small-sample distribution of the maximum conditional likelihood estimator is strongly skewed to the right, which may produce Wald-type confidence intervals with lower limits that are less than the number of captured individuals or even negative.

In this talk, we present a full likelihood approach based on Huggins and Alho's model. We show that the null distribution of the empirical likelihood ratio for the abundance is asymptotically chi-square with one degree of freedom, and the maximum empirical likelihood estimator achieves semiparametric efficiency. We further propose an expectation–maximization algorithm to numerically calculate the proposed point estimate and empirical likelihood ratio function. Simulation studies show that the empirical-likelihood-based method is superior to the conditional-likelihood-based method: its confidence interval has much better coverage, and the maximum empirical likelihood estimator has a smaller mean square error.

September 23, 2022 from 15:30 to 16:30 (Montreal/EST time)

Colloquium presented by **Anush Tserunyan (McGill University)**

Pointwise ergodic theorems provide a bridge between the global behaviour of the dynamical system and the local combinatorial statistics of the system at a point. Such theorem have been proven in different contexts, but typically for actions of semigroups on a probability space. Dating back to Birkhoff (1931), the first known pointwise ergodic theorem states that for a measure-preserving ergodic transformation T on a probability space, the mean of a function (its global average) can be approximated by taking local averages of the function at a point x over finite sets in the forward-orbit of x, namely {x, Tx, ..., T^n x}. Almost a century later, we revisit Birkhoff's theorem and turn it backwards, showing that the averages along trees of possible pasts also approximate the global average. This backward theorem for a single transformation surprisingly has applications to actions of free groups, which we will also discuss. This is joint work with Jenna Zomback.

**Address**

September 16, 2022 from 15:30 to 16:30 (Montreal/EST time)

Colloquium presented by **Chandrashekhar Khare (UCLA)**

Ramanujan made a series of influential conjectures in his 1916 paper ``On some arithmetical functions’' on what is now called the Ramanujan τ\tauτ function. A congruence Ramanujan observed for τ(n)\tau(n)τ(n) modulo 691 in the paper led to Serre and Swinnerton-Dyer developing a geometric theory of mod ppp modular forms. It was in the context of the theory of mod ppp modular forms that Serre made his modularity conjecture, which was initially formulated in a letter of Serre to Tate in 1973.

I will describe the path from Ramanujan's work in 1916, to the formulation of a first version of Serre's conjecture in 1973, to its resolution in 2009 by Jean-Pierre Wintenberger and myself. I will also try to indicate why this subject is very much alive and, in spite of all the progress, still in its infancy.

**Address**

September 2, 2022 from 15:30 to 16:30 (Montreal/EST time)

Colloquium presented by **Véronique Bazier-Matte (Université Laval)**

Les algèbres amassées sont des algèbres de polynômes de Laurent dont les générateurs s’obtiennent par un processus récursif appelé mutation. On commence avec une graine, paire formée d’un ensemble de n variables, appelé amas, et d’un graphe orienté à n points. La mutation d’une graine remplace une variable à la fois et modifie le graphe, donnant ainsi une nouvelle graine. L’algèbre amassée est engendrée par toutes les variables obtenues par mutations successives, qu’on appelle variables amassées.

Les algèbres amassées ont été définies il y a 20 ans, et depuis, des liens entre elles et divers champs de recherche ont été découverts. Récemment, avec mon collaborateur, nous avons établi un lien entre la théorie des noeuds et les algèbres amassées. Cet exposé introduira d'abord les algèbres amassées puis présentera la connection entre ces dernières et le polynôme d'Alexander d'un noeud.

**Address**

May 20, 2022 from 15:30 to 16:30 (Montreal/EST time)

Colloquium presented by **Isabelle Gallagher (Ecole Normale Supérieure)**

2022 Aisenstadt Chair recipient

The evolution of a gas can be described by different models depending on the observation scale. A natural question, raised by Hilbert in his sixth problem, is whether these models provide consistent predictions. In particular, for rarefied gases, it is expected that continuum laws of kinetic theory can be obtained directly from molecular dynamics governed by the fundamental principles of mechanics. In the case of hard sphere gases, Lanford showed in 1975 that the Boltzmann equation emerges as the law of large numbers in the low density limit, at least for very short times. The goal of this talk is to explain the heuristics of his proof and present recent progress in the understanding of this limiting process.

May 6, 2022 from 15:30 to 16:30 (Montreal/EST time)

Colloquium presented by **Slawomir Solecki (Cornell University)**

The behavior of a measure preserving transformation, even a generic one, is highly non-uniform. In contrast to this observation, a different picture of a very uniform behavior of the closed group generated by a generic measure preserving transformation has emerged. This picture included substantial evidence that pointed to these groups being all topologically isomorphic to a single group, namely, $L^0$---the non-locally compact, topological group of all Lebesgue measurable functions from $[0,1]$ to the circle. In fact, Glasner and Weiss asked if this was the case.

We will describe the background touched on above, including the connections with Descriptive Set Theory. Further, we will indicate a proof of the following theorem that answers the Glasner--Weiss question in the negative: for a generic measure preserving transformation $T$, the closed group generated by $T$ is {\bf not} topologically isomorphic to $L^0$.

April 29, 2022 from 15:30 to 16:30 (Montreal/EST time) Zoom meeting

Colloquium presented by **Forrest Crawford (Yale University)**

Simple mathematical models of COVID-19 transmission gained prominence in the early days of the pandemic. These models provided researchers and policymakers with qualitative insight into the dynamics of transmission and quantitative predictions of disease incidence. More sophisticated models incorporated new information about the natural history of COVID-19 disease and the interaction of infected individuals with the healthcare system, to predict diagnosed cases, hospitalization, ventilator usage, and death. Models also provided intuition for discussions about outbreaks, vaccination, and the effects of non-pharmaceutical interventions like social distancing guidelines and stay-at-home orders. But as the pandemic progressed, complex real-world interventions took effect, people everywhere changed their behavior, and the usefulness of simple mathematical models of COVID-19 transmission diminished. This challenge forced researchers to think more broadly about empirical data sources that could help predictive models regain their utility for guiding public policy. In this presentation, I will describe my view of the successes and failures of population-level transmission models in the context of the COVID-19 pandemic. I will outline the evolution of a project to predict COVID-19 incidence in the state of Connecticut, from development of a transmission model to engagement with public health policymakers and initiation of a new data collection effort. I argue that a new data source – passive measurement of close interpersonal contact via mobile device location data – is a promising way to overcome many of the shortcomings of traditional transmission models. I conclude with a summary of the impact this work has had on the COVID-19 response in Connecticut and beyond.

April 22, 2022 from 15:30 to 16:30 (Montreal/EST time)

Colloquium presented by **Joel Kamnitzer (McGill University)**

The cactus group is a cousin of the braid group and shares many of its beautiful properties. It is the fundamental group of the moduli space of points on RP^1. It also acts on many collections of combinatorial objects. I will explain how we use the cactus group to understand monodromy of eigenvectors for Gaudin algebras.

April 15, 2022 from 15:00 to 16:00 (Montreal/EST time)

Colloquium presented by **Pierre Cardaliaguet (Université Paris-Dauphine)**

Mean Field Game is the study of the dynamical behavior of a large number of agents in interaction. For instance, it can model be the dynamics of a crowd, or the production of a renewable resource by a large amount of producers. The analysis of these models, first introduced in the economic literature under the terminology of “heterogenous agent models, has known a spectacular development with the pioneering woks of Lasry and Lions and of Caines, Huang and Malhamé. The aim of the talk will be to illustrate the theory through a few models and present some of the main results and open questions.

April 8, 2022 from 15:30 to 16:30 (Montreal/EST time) Zoom meeting

Colloquium presented by **Robert Raussendorf (UBC)**

2021 CAP-CRM Prize Recipient

We show that every quantum computation can be described by a probabilistic update of a probability distribution on a finite phase space. Negativity in a quasiprobability function is not required in states or operations, which is a very unusual feature. Nonetheless, our result is consistent with Gleason’s Theorem and the Pusey-Barrett- Rudolph theorem.

The reason I have chosen this subject for my talk is two-fold: (i) It gives the audience a glimpse of the quest to understand the quantum mechanical cause for speed-up in quantum computation, which is one of the central questions on the theory side of the field, and (ii) Maybe there can be feedback from the audience. The structures underlying the above probabilistic model are the so-called Lambda-polytopes, which are highly symmetric objects. At present we only know very few general facts about them. Help with analysing them would be appreciated!

Joint work with Michael Zurel and Cihan Okay,

Journal reference: Phys. Rev. Lett. 125, 260404 (2020)

April 1, 2022 from 14:00 to 15:00 (Montreal/EST time) Zoom meeting

Colloquium presented by **Sibylle Schroll (University of Cologne)**

Derived categories are in general not easy to parse. However, in certain cases, combinatorial models give a good picture of these categories. One such case are the bounded derived categories of gentle algebras which can be represented in terms of curves and crossings of curves on surfaces. In this talk, we will give the construction of these surface models and briefly explain how they are connected to the homological mirror symmetry programme. We will show how a combination of surface combinatorics and representation theory can give new insights into the associated categories.

March 25, 2022 from 15:30 to 16:30 (Montreal/EST time) Zoom meeting

Colloquium presented by **Heather Macbeth (Fordham university)**

In the last thirty years, computer proof verification became a mature technology, with successes including the checking of the Four-Colour Theorem, the Odd Order Theorem, and Hales' proof of the Kepler Conjecture. Recent advances such as the "Liquid Tensor Experiment" verifying a recent theorem of Scholze have provided further momentum, as likewise have promising experiments integrating this technology with machine learning.

I will briefly describe some of these developments. I will then try to describe, more generally, what it *feels* like to carry out research-level computer verifications of mathematics proofs: the level of expression one has access to, the ways one finds oneself interrogating and reorganizing a paper proof, the kinds of arguments which are more tedious (or less tedious!) than on paper.

March 18, 2022 from 15:30 to 16:30 (Montreal/EST time)

Colloquium presented by **Bernard Derrida (École Normale Supérieure)**

2022 Aisenstadt Chair recipient

Statistical Physics allowed to unify, at the end of the 19th century, Newton's mechanics and thermodynamics. It gave a way to predict the amplitude of fluctuations around the physical laws which were known at that time. Einstein, in his very first works, showed that the measurement of these fluctuations allowed to estimate the size of atoms. His reasoning, which was at the origin of the linear response theory, applied to the black body gave one of the first evidences of the duality wave-particle in Quantum Mechanics. Statistical Physics gives also a framework to predict large deviations for systems at equilibrium. In the last two decades, major efforts were devoted to extend our understanding of the statistical laws of fluctuations and large deviations to non-equilibrium systems. This talk will try to present some of the recent progresses.

March 11, 2022 from 15:30 to 16:30 (Montreal/EST time) Zoom meeting

Colloquium presented by **Evgeny Gorksy (UC Davis)**

Khovanov and Rozansky defined in 2005 a triply graded link homology theory which generalizes HOMFLY-PT polynomial. In this talk, I will outline some known results and structures in Khovanov-Rozansky homology, describe its connection to q,t-Catalan combinatorics and present several geometric models for some classes of links.

February 18, 2022 from 15:30 to 16:30 (Montreal/EST time) Zoom meeting

Colloquium presented by **Stanislav Volgushev (University of Toronto)**

Extremal graphical models are sparse statistical models for multivariate extreme events. The underlying graph encodes conditional independencies and enables a visual interpretation of the complex extremal dependence structure. For the important case of tree models, we provide a data-driven methodology for learning the graphical structure. We show that sample versions of the extremal correlation and a new summary statistic, which we call the extremal variogram, can be used as weights for a minimum spanning tree to consistently recover the true underlying tree. Remarkably, this implies that extremal tree models can be learned in a completely non-parametric fashion by using simple summary statistics and without the need to assume discrete distributions, existence of densities, or parametric models for marginal or bivariate distributions. Extensions to more general graphs are also discussed.

February 11, 2022 from 15:30 to 16:30 (Montreal/EST time) Zoom meeting

Colloquium presented by **Ryan Hynd (University of Pennsylvania)**

I will discuss the time evolution of a collection of particles that interact primarily through perfectly inelastic collisions. I will explain why this problem is tractable if the particles are constrained to lie on a line versus if they are allowed to move freely in space. In particular, I'll also describe an equation at the heart of this difficulty which some researchers believe has been solved and others do not. This topic has motivations in astronomy and connections with optimal mass transportation which I will touch upon if time permits.

February 4, 2022 from 12:00 to 13:00 (Montreal/EST time) Zoom meeting

Colloquium presented by **Sarah Zerbes (ETH Zürich)**

L-functions are one of the central objects of study in number theory. There are many beautiful theorems and many more open conjectures linking their values to arithmetic problems. The most famous example is the conjecture of Birch and Swinnerton-Dyer, which is one of the Clay Millenium Prize Problems. I will discuss this conjecture and some related open problems, and I will describe some recent progress on these conjectures, using tools called `Euler systems’.

January 28, 2022 from 15:30 to 16:30 (Montreal/EST time) Zoom meeting

Colloquium presented by **Gilles Stupfler (ENSAI)**

Statistical risk assessment, in particular in finance and insurance, requires estimating simple indicators to summarize the risk incurred in a given situation. Of most interest is to infer extreme levels of risk so as to be able to manage high-impact rare events such as extreme climate episodes or stock market crashes. A standard procedure in this context, whether in the academic, industrial or regulatory circles, is to estimate a well-chosen single quantile (or Value-at-Risk). One drawback of quantiles is that they only take into account the frequency of an extreme event, and in particular do not give an idea of what the typical magnitude of such an event would be. Another issue is that they do not induce a coherent risk measure, which is a serious concern in actuarial and financial applications. In this talk, after giving a leisurely tour of extreme quantile estimation, I will explain how, starting from the formulation of a quantile as the solution of an optimization problem, one may come up with two alternative families of risk measures, called expectiles and extremiles, in order to address these two drawbacks. I will give a broad overview of their properties, as well as of their estimation at extreme levels in heavy-tailed models, and explain why they constitute sensible alternatives for risk assessment using real data applications. This is based on joint work with Abdelaati Daouia, Irène Gijbels, Stéphane Girard, Simone Padoan and Antoine Usseglio-Carleve.

**Address**

January 21, 2022 from 14:00 to 15:00 (Montreal/EST time) Zoom meeting

Colloquium presented by **Allen Knutson (Cornell University)**

Nobody knows whether the scheme "pairs of commuting nxn matrices" is reduced. I'll show how this scheme relates to matrix Schubert varieties, and give a formula for its equivariant cohomology class (and that of many other varieties) using "generic pipe dreams" that I'll introduce. These interpolate between ordinary and bumpless pipe dreams. With those, I'll rederive both formulae (ordinary and bumpless) for double Schubert polynomials. This work is joint with Paul Zinn-Justin.

**Address**

January 14, 2022 from 11:00 to 12:00 (Montreal/EST time) Zoom meeting

Colloquium presented by **Eva Miranda (Polytechnic University of Catalonia, Spain)**

What physical systems can be non-computational? (Roger Penrose, 1989). Is hydrodynamics capable of calculations? (Cris Moore, 1991). Can a mechanical system (including the trajectory of a fluid) simulate a universal Turing machine? (Terence Tao, 2017).

The movement of an incompressible fluid without viscosity is governed by Euler equations. Its viscid analogue is given by the Navier-Stokes equations whose regularity is one of the open problems in the list of problems for the Millenium by

the Clay Foundation. The trajectories of a fluid are complex. Can we measure its levels of complexity (computational, logical and dynamical)?

In this talk, we will address these questions. In particular, we will show how to construct a 3-dimensional Euler flow which is Turing complete. Undecidability of fluid paths is then a consequence of the classical undecidability of the halting

problem proved by Alan Turing back in 1936. This is another manifestation of complexity in hydrodynamics which is very different from the theory of chaos.

Our solution of Euler equations corresponds to a stationary solution or Beltrami field. To address this problem, we will use a mirror [5] reflecting Beltrami fields as Reeb vector fields of a contact

structure. Thus, our solutions import techniques from geometry to solve a problem in fluid dynamics. But how general are Euler flows? Can we represent any dynamics as an Euler flow? We will address this universality problem using the Beltrami/Reeb mirror again and Gromov's h-principle. We will also consider the non-stationary case. These universality features illustrate the complexity of Euler flows. However, this construction is not "physical" in the sense that the associated metric is not the euclidean metric. We will announce an euclidean construction and its implications to complexity and undecidability.

These constructions [1,2,3,4] are motivated by Tao's approach to the problem of Navier-Stokes [7,8,9] which we will also explain.

[1] R. Cardona, E. Miranda, D. Peralta-Salas, F. Presas. Universality of Euler flows and flexibility of Reeb

embeddings. https://arxiv.org/abs/1911.01963.

[2] R. Cardona, E. Miranda, D. Peralta-Salas, F. Presas. Constructing Turing complete Euler flows in

dimension 3. Proc. Natl. Acad. Sci. 118 (2021) e2026818118.

[3] R. Cardona, E. Miranda, D. Peralta-Salas. Turing universality of the incompressible Euler equations

and a conjecture of Moore. Int. Math. Res. Notices, , 2021;, rnab233,

https://doi.org/10.1093/imrn/rnab233

[4] R. Cardona, E. Miranda, D. Peralta-Salas. Computability and Beltrami fields in Euclidean space.

https://arxiv.org/abs/2111.03559

[5] J. Etnyre, R. Ghrist. Contact topology and hydrodynamics I. Beltrami fields and the Seifert conjecture.

Nonlinearity 13 (2000) 441–458.

[6] C. Moore. Generalized shifts: unpredictability and undecidability in dynamical systems. Nonlinearity

4 (1991) 199–230.

[7] T. Tao. On the universality of potential well dynamics. Dyn. PDE 14 (2017) 219–238.

[8] T. Tao. On the universality of the incompressible Euler equation on compact manifolds. Discrete

Cont. Dyn. Sys. A 38 (2018) 1553–1565.

[9] T. Tao. Searching for singularities in the Navier-Stokes equations. Nature Rev. Phys. 1 (2019) 418–419.

**Address**

December 17, 2021 from 10:00 to 11:00 (Montreal/EST time) Zoom meeting

Colloquium presented by **Yen-Tsung Huang (Institute of Statistical Science, Academia Sinica)**

A causal mediation model with multiple time-to-event mediators is exemplified by the natural course of human disease marked by sequential milestones with a time-to-event nature. For example, from hepatitis B infection to death, patients may experience intermediate events such as liver cirrhosis and liver cancer. The sequential events of hepatitis, cirrhosis, cancer, and death are susceptible to right censoring; moreover, the latter events may preclude the former events. Casting the natural course of human diseases in the framework of causal mediation modeling, we establish a model with intermediate and terminal events as the mediators and outcomes, respectively. We define the interventional analog of path-specific effects (iPSEs) as the effect of an exposure on a terminal event mediated (or not mediated) by any combination of intermediate events without parametric models. The expression of a counting process-based counterfactual hazard is derived under the sequential ignorability assumption. We employ composite nonparametric likelihood estimation to obtain maximum likelihood estimators for the counterfactual hazard and iPSEs. Our proposed estimators achieve asymptotic unbiasedness, uniform consistency, and weak convergence. Applying the proposed method, we show that hepatitis B induced mortality is mostly mediated through liver cancer and/or cirrhosis whereas hepatitis C induced mortality may be through extrahepatic diseases.

**Address**

December 10, 2021 from 14:00 to 15:00 (Montreal/EST time) On location

Colloquium presented by **Samit Dasgupta (Duke University)**

In this talk we will discuss two central problems in algebraic number theory and their interconnections: explicit class field theory and the special values of L-functions. The goal of explicit class field theory is to describe the abelian extensions of a ground number field via analytic means intrinsic to the ground field; this question lies at the core of Hilbert's 12th Problem. Meanwhile, there is an abundance of conjectures on the values of L-functions at certain special points. Of these, Stark's Conjecture has relevance toward explicit class field theory. I will describe two recent joint results with Mahesh Kakde on these topics. The first is a proof of the Brumer-Stark conjecture away from p=2. This conjecture states the existence of certain canonical elements in abelian extensions of totally real fields. The second is a proof of an exact formula for Brumer-Stark units that has been developed over the last 15 years. We show that these units together with other easily written explicit elements generate the maximal abelian extension of a totally real field, thereby giving a p-adic solution to the question of explicit class field theory for these fields.

**Address**

December 3, 2021 from 15:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **Valentino Tosatti (McGill University)**

K3 surfaces are a class of compact complex manifolds that enjoys many special properties and play an important role in several areas of mathematics. In this colloquium I will discuss a new interplay between complex geometry and analysis on K3 surfaces equipped with their Calabi-Yau metrics, and dynamics of holomorphic diffeomorphisms of these surfaces, that Simion Filip and I have been investigating recently.

**Address**

November 26, 2021 from 15:30 to 16:30 (Montreal/EST time) Zoom meeting

Colloquium presented by **Simon Bonner (University of Western Ontario)**

Monitoring marked individuals is a common strategy in studies of wild animals (referred to as mark-recapture or capture-recapture experiments) and hard to track human populations (referred to as multi-list methods or multiple-systems estimation). A standard assumption of these techniques is that individuals can be identified uniquely and without error, but this can be violated in many ways. In some cases, it may not be possible to identify individuals uniquely because of the study design or the choice of marks. Other times, errors may occur so that individuals are incorrectly identified. I will discuss work with my collaborators over the past 10 years developing methods to account for problems that arise when are only individuals are only partially identified. I will present theoretical aspects of this research, including an introduction to the latent multinomial model and algebraic statistics, and also describe applications to studies of species ranging from the golden mantella (an endangered frog endemic to Madagascar measuring only 20 mm) to the whale shark (the largest known species of sh measuring up to 19 m).

**Address**

November 19, 2021 from 15:30 to 16:30 (Montreal/EST time) Zoom meeting

Colloquium presented by **Tristan Collins (MIT)**

2021 André Aisenstadt Prize in Mathematics Recipient

A fundamental problem in string theory is the multitude of distinct geometries which give rise to consistent solutions of the vacuum equations of motion.

One possible resolution of this "vacuum degeneracy" problem is the "fantasy" that the moduli space of string vacua is connected through the process of "geometric transitions".

I will discuss some geometric problems associated to this fantasy and their applications.

**Address**

November 12, 2021 from 15:30 to 16:30 (Montreal/EST time)

Colloquium presented by **Gabor Lugosi (ICREA-UPF and BSE)**

One of the most basic problems in statistics is the estimation of the mean of a random vector, based on independent observations. This problem has received renewed attention in the last few years, both from statistical and computational points of view. In this talk, we review some recent results on the statistical performance of mean estimators that allow heavy tails and adversarial contamination in the data. In particular, we are interested in estimators that have a near-optimal error in all directions in which the variance of the one dimensional marginal of the random vector is not too small. The material of this talk is based on a series of joint papers with Shahar Mendelson.

November 5, 2021 from 15:30 to 16:30 (Montreal/EST time)

Colloquium presented by **Louise Poirier (Université de Montréal)**

The presentation is divided in two parts. First, the main results of our study on the “mathematical portrait” of Québec students carried out as part of the En avant math! Project (CRM-CIRANO joint project supported by the Ministry of Finance) will be presented. This report is based on the one hand on the results of the international TIMS and PISA tests for Québec elementary and secondary school students and on the other hand, on the situation of mathematics in Québec universities based on data from the Bureau de coopération interuniversitaire (BCI) in terms of evolution of student enrollments and portrait of students ( gender and status). BCI data shows that the typical student enrolled in math at Québec universities is a Canadian citizen, white and male. And the total number of registrants decreases each year (except, perhaps for the PhD students). Where are the girls? Where are the students from recent immigration? And yet on the PISA and TIMMS tests, immigrant students perform better in Canada than Canadian students (the opposite is true for most of the OECD countries).

Then, in the light of the results of the student portrait, we will discuss the social issues for more inclusive mathematics. Collaborative research with the Inuit communities of Nunavik will illustrate our point.

Note: Denis Gued will excuse me for borrowing this title. These words are reflected in his interview "Let's make mathematics amiable" with L’express.

https://www.lexpress.fr/informations/rendons-les-mathematiques-aimables_640643.html

October 29, 2021 from 15:30 to 16:30 (Montreal/EST time)

Colloquium presented by **Tiffany Timbers (University of British Columbia)**

In the data science courses at the University of British Columbia, we define data science as the study, development and practice of reproducible and auditable processes to obtain insight from data. While reproducibility is core to our definition, most data science learners enter the field with other aspects of data science in mind, for example predictive modelling, which is often one of the most interesting topic to novices. This fact, along with the highly technical nature of the industry standard reproducibility tools currently employed in data science, present out-ofthe gate challenges in teaching reproducibility in the data science classroom. Put simply, students are not as intrinsically motivated to learn this topic, and it is not an easy one for them to learn. What can a data science educator do? Over several iterations of teaching courses focused on reproducible data science tools and workflows, we have found that providing extra motivation, guided instruction and lots of practice are key to effectively teaching this challenging, yet important subject. Here we present examples of how we deeply motivate, effectively guide and provide ample practice opportunities to data science students to effectively engage them in learning about this topic.

October 15, 2021 from 15:30 to 16:30 (Montreal/EST time)

Colloquium presented by **Giulio Tiozzo (University of Toronto)**

2021 André Aisenstadt Prize in Mathematics Recipient

The notion of topological entropy, arising from information theory, is a fundamental tool to understand the complexity of a dynamical system. When the dynamical system varies in a family, the natural question arises of how the entropy changes with the parameter.

In the last decade, W. Thurston introduced these ideas in the context of complex dynamics by defining the "core entropy" of a quadratic polynomials as the entropy of a certain forward-invariant set of the Julia set (the Hubbard tree).

As we shall see, the core entropy is a purely topological/combinatorial quantity which nonetheless captures the richness of the fractal structure of the Mandelbrot set. In particular, we will relate the variation of such a function to the geometry of the Mandelbrot set. We will also prove that the core entropy on the space of polynomials of a given degree varies continuously, answering a question of Thurston.

Finally, we will provide a new interpretation of core entropy in terms of measured laminations, and discuss its finer regularity properties such as its Holder exponent.

October 8, 2021 from 11:00 to 12:00 (Montreal/EST time)

Colloquium presented by **Yuansi Chen (Duke University)**

In recent work, Chen (2020) improved Eldan's stochastic localization proof technique, which was deployed in Lee and Vempala (2017), to prove an almost constant Cheeger isoperimetric coefficient in the KLS conjecture with dimension dependency d^o(1). Consequently, his proof also provides a substantial advance toward the resolution of Bourgain's slicing conjecture and the thin-shell conjecture. After getting conformable with Eldan's stochastic localization technique, in this talk we navigate through how to refine the technique to provide the current best bound. We will complete the self-contained proof of Chen (2020) and highlight the new ideas involved. Finally, we will discuss some extensions and provide an outlook for future research directions.

**Address**

October 1, 2021 from 11:00 to 12:00 (Montreal/EST time)

Colloquium presented by **Bo'az Klartag (Weizmann Institute of Science)**

High-dimensional problems with a geometric flavor appear in a number of branches of mathematics and mathematical physics. A priori, it seems that the immense diversity observed in high dimensions would make it impossible to formulate general, interesting theorems that apply to large classes of high-dimensional geometric objects. In this talk we will discuss situations in which high dimensionality, when viewed correctly, induces remarkable order and simplicity rather than complication. For example, Dvoretzky's theorem demonstrates that any high-dimensional convex body possesses nearly-Euclidean sections of large dimension. Another example is the central limit theorem for convex bodies, according to which any high-dimensional convex body has approximately-Gaussian marginals. There are strong motifs in high-dimensional geometry, such as the concentration of measure, which appear to compensate for the large number of different configurations. Convexity allows us to harness these motifs in order to formulate elegant and non-trivial theorems.

**Address**

September 24, 2021 from 15:00 to 16:00 (Montreal/EST time)

Colloquium presented by **Jennifer Hill (NYU Steinhardt)**

Most researchers in the social, behavioral, and health sciences are taught to be extremely cautious in making causal claims. However, causal inference is a necessary goal in research for addressing many of the most pressing questions around policy and practice. In the past decade, causal methodologists have increasingly been using and touting the benefits of more complicated machine learning algorithms to estimate causal effects. These methods can take some of the guesswork out of analyses, decrease the opportunity for “p-hacking,” and may be better suited for more fine-tuned tasks such as identifying varying treatment effects and generalizing results from one population to another. However, should these more advanced methods change our fundamental views about how difficult it is to infer causality? In this talk I will discuss some potential advantages and disadvantages of using machine learning for causal inference and emphasize ways that we can all be more transparent in our inferences and honest about their limitations.

April 30, 2021 from 15:00 to 16:00 (Montreal/EST time)

Colloquium presented by **Eva Bayer-Fluckiger (École Polytechnique Fédérale de Lausanne)**

After a historical introduction to knot theory, the talk will be centered around two knot invariants, the Alexander polynomial and the signature. The aim is to introduce a finite abelian group that controls their relationship, and to illustrate this by several examples. Using Seifert matrices, the geometric questions are translated into arithmetic ones.

April 23, 2021 from 15:00 to 16:00 (Montreal/EST time)

Colloquium presented by **Adrian Lewis (Cornell University)**

To the dismay and irritation of the variational analysis community, practitioners of deep learning often implement gradient-based optimization via automatic differentiation and blithely apply the result to nonsmooth objectives. Worse, they then gleefully point out numerical convergence. In fact, as elegantly remarked by Bolte and Pauwels, automatic differentiation produces a novel generalized gradient: a conservative field with enough calculus to prove convergence of stochastic subgradient descent, as practiced in deep learning. I will sketch this interplay of analytic and algorithmic ideas, and explain how, for concrete objectives (typically semi-algebraic), this novel generalized gradient just slightly modifies Clarke's original notion.

Joint work with Tonghua Tian.

April 16, 2021 from 15:00 to 16:00 (Montreal/EST time)

Colloquium presented by **Kilian Raschel (Université de Tours)**

We consider a reflected Brownian motion in a two-dimensional wedge. Under standard assumptions on the parameters of the model (opening of the wedge, angles of the reflections on the axes, drift), we study the algebraic and differential nature of the Laplace transform of its stationary distribution. We derive necessary and sufficient conditions for this Laplace transform to be rational, algebraic, differentially finite or more generally differentially algebraic. These conditions are explicit linear dependencies among the angles involved in the definition of the model.

To prove these results, we start from a functional equation that the Laplace transform satisfies, to which we apply tools from diverse horizons. To establish differential algebraicity, a key ingredient is Tutte's invariant approach, which originates in enumerative combinatorics. To establish differential transcendence, we turn the functional equation into a difference equation and apply Galoisian results on the nature of the solutions to such equations.

This is a joint work with M. Bousquet-Mélou, A. Elvey Price, S. Franceschi and C. Hardouin (https://arxiv.org/abs/2101.01562).

April 9, 2021 from 15:00 to 16:00 (Montreal/EST time)

Colloquium presented by **Jane Wang (Cornell University)**

Why do animals move the way they do? Bacteria, insects, birds, and fish share with us the necessity to move so as to live. Although each organism follows its own evolutionary course, it also obeys a set of common laws. At the very least, the movement of animals, like that of planets, is governed by Newton’s law: All things fall. On Earth, most things fall in air or water, and their motions are thus subject to the laws of hydrodynamics. Through trial and error, animals have found ways to interact with fluid so they can float, drift, swim, sail, glide, soar, and fly. This elementary struggle to escape the fate of falling shapes the development of motors, sensors, and mind. Perhaps we can deduce parts of their neural computations by understanding what animals must do so as not to fall.

We have been seeking mechanistic explanations of the complex movement of insect flight. Starting from the Navier-Stokes equations governing the unsteady aerodynamics of flapping flight, we worked to build a theoretical framework for computing flight and for studying the control of flight. I will discuss our recent computational and experimental studies of the balancing act of dragonflies and fruit flies: how a dragonfly recovers from falling upside-down and how a fly balances in air. In each case, the physics of flight informs us about the neural feedback circuitries underlying their fast reflexes.

March 19, 2021 from 15:00 to 16:00 (Montreal/EST time)

Colloquium presented by **François Dubois (Le CNAM - Paris, Member of IRL-CRM CNRS)**

We first recall some elements of history of the construction of lattice Boltzmann schemes. Then we present our "ABCD" approach, founded on the property that the numerical scheme is exact for the advection equation with the velocities of the lattice. This asymptotic analysis allows to write at several orders the conservative partial differential equations equivalent to the numerical scheme. A fit of parameters permits in favorable cases a precise approximation of compressible fluids equations.

March 12, 2021 from 15:30 to 16:30 (Montreal/EST time)

Colloquium presented by **Jay Breidt (Colorado State University, USA)**

Informative selection, in which the distribution of response variables given that they are sampled is different from their distribution in the population, is pervasive in complex surveys. Failing to take such informativeness into account can produce severe inferential errors, including biased and inconsistent estimation of population parameters. While several parametric procedures exist to test for informative selection, these methods are limited in scope and their parametric assumptions are difficult to assess. We consider two classes of nonparametric tests of informative selection. The first class is motivated by classic nonparametric two-sample tests. We compare weighted and unweighted empirical distribution functions and obtain tests for informative selection that are analogous to Kolmogorov-Smirnov and Cramer-von Mises. For the second class of tests, we adapt a kernel-based learning method that compares distributions based on their maximum mean discrepancy. The asymptotic distributions of the test statistics are established under the null hypothesis of noninformative selection. Simulation results show that our tests have power competitive with existing parametric tests in a correctly specified parametric setting, and better than those tests under model misspecification. A recreational angling application illustrates the methodology.

This is joint work with Teng Liu, Colorado State University.

February 26, 2021 from 15:00 to 16:00 (Montreal/EST time)

Colloquium presented by **Jean-Pierre Demailly (Université Grenoble Alpes, France)**

A projective algebraic variety is defined as the zero locus of a finite family of homogeneous polynomials. Over the field of complex numbers, the geometry of such varieties is governed to a large extent by the sign, in a suitable sense, of the Ricci curvature form. When this sign is negative, the variety is expected to exhibit certain hyperbolicity properties in the sense of Kobayashi - as well as further very deep number-theoretic properties that are mostly conjectural, in the arithmetic situation. In particular, all entire holomorphic curves drawn on it should be contained in a proper algebraic subvariety: this is a famous conjecture of Green-Griffiths and Lang. Following recent ideas of D. Brotbek, we will try to explain here a rather elementary proof of a related conjecture of Kobayashi, stating that a general algebraic hypersurface of sufficiently high degree is hyperbolic, i.e. does not contain any entire holomorphic curve.

February 19, 2021 from 15:30 to 16:30 (Montreal/EST time)

Colloquium presented by **Larry Guth (MIT)**

The local smoothing problem asks about how much solutions to the wave equation can focus. It was formulated by Chris Sogge in the early 90s. Hong Wang, Ruixiang Zhang, and I recently proved the conjecture in two dimensions.

February 12, 2021 from 15:30 to 16:30 (Montreal/EST time)

Colloquium presented by **Mikael Kuusela (Carnegie Mellon University, USA)**

Tropical cyclones (TCs), driven by heat exchange between the air and sea, pose a substantial risk to many communities around the world. Accurate characterization of the subsurface ocean thermal response to TC passage is crucial for accurate TC intensity forecasts and for understanding the role TCs play in the global climate system, yet that characterization is complicated by the high-noise ocean environment, correlations inherent in spatio-temporal data, relative scarcity of in situ observations and the entanglement of the TC-induced signal with seasonal signals. We present a general methodological framework that addresses these difficulties, integrating existing techniques in seasonal mean field estimation, Gaussian process modeling, and nonparametric regression into a functional ANOVA model. Importantly, we improve upon past work by properly handling seasonality, providing rigorous uncertainty quantification, and treating time as a continuous variable, rather than producing estimates that are binned in time. This functional ANOVA model is estimated using in situ subsurface temperature profiles from the Argo fleet of autonomous floats through a multi-step procedure, which (1) characterizes the upper ocean seasonal shift during the TC season; (2) models the variability in the temperature observations; (3) fits a thin plate spline using the variability estimates to account for heteroskedasticity and correlation between the observations. This spline fit reveals the ocean thermal response to TC passage. Through this framework, we obtain new scientific insights into the interaction between TCs and the ocean on a global scale, including a three-dimensional characterization of the near-surface and subsurface cooling along the TC storm track and the mixing-induced subsurface warming on the track's right side. Joint work with Addison Hu, Ann Lee, Donata Giglio and Kimberly Wood.

February 5, 2021 from 15:00 to 16:00 (Montreal/EST time)

Colloquium presented by **Egor Shelukhin (Université de Montréal, Canada)**

2020 André Aisenstadt Prize in Mathematics Recipient

In the early 60s Arnol'd has conjectured that Hamiltonian diffeomorphisms, the motions of classical mechanics, often possess more fixed points than required by classical topological considerations. In the late 80s and early 90s Floer has developed a powerful theory to approach this conjecture, considering fixed points as critical points of a certain functional. Recently, in joint work with L. Polterovich, we observed that Floer theory filtered by the values of this functional fits into the framework of persistence modules and their barcodes, originating in data sciences. I will review these developments and their applications, which arise from a natural time-symmetry of Hamiltonians. This includes new constraints on one-parameter subgroups of Hamiltonian diffeomorphisms, as well as my recent solution of the Hofer-Zehnder periodic points conjecture. The latter combines barcodes with equivariant cohomological operations in Floer theory recently introduced by Seidel to form a new method with further consequences.

January 29, 2021 from 15:30 to 16:30 (Montreal/EST time)

The under-five mortality rate (U5MR) is a key barometer of the health of a nation. Unfortunately, many people living in low- and middle-income countries are not covered by civil registration systems. This makes estimation of the U5MR, particularly at the subnational level, difficult. In this talk, I will describe models that have been developed to produce the official United Nations (UN) subnational U5MR estimates in 22 countries. Estimation is based on household surveys, which use stratified, two-stage cluster sampling. I will describe a range of area- and unit-level models and describe the rationale for the modeling we carry out. Data sparsity in time and space is a key challenge, and smoothing models are vital. I will discuss the advantages and disadvantages of discrete and continuous spatial models, in the context of estimation at the scale at which health interventions are made. Other issues that will be touched upon include: design-based versus model-based inference; adjustments for HIV epidemics; the inclusion of so-called indirect (summary birth history) data; reproducibility through software availability; benchmarking; how to deal with incomplete geographical data; and working with the UN to produce estimates.

January 22, 2021 from 15:00 to 16:00 (Montreal/EST time)

Colloquium presented by **Robert Haslhofer (University of Toronto, Canada)**

2020 André Aisenstadt Prize in Mathematics Recipient

A family of surfaces moves by mean curvature flow if the velocity at each point is given by the mean curvature vector. Mean curvature flow first arose as a model of evolving interfaces and has been extensively studied over the last 40 years.

In this talk, I will give an introduction and overview for a general mathematical audience. To gain some intuition we will first consider the one-dimensional case of evolving curves. We will then discuss Huisken’s classical result that the flow of convex surfaces always converges to a round point. On the other hand, if the initial surface is not convex we will see that the flow typically encounters singularities. Getting a hold of these singularities is crucial for most striking applications in geometry, topology and physics. Specifically, singularities can be either of neck-type or conical-type. We will discuss examples from the 90s, which show, both experimentally and theoretically, that flow through conical singularities is utterly non-unique.

In the last part of the talk, I will report on recent work with Kyeongsu Choi, Or Hershkovits and Brian White, where we proved that mean curvature flow through neck-singularities is unique. The key for this is a classification result for ancient asymptotically cylindrical flows that describes all possible blowup limits near a neck-singularity. In particular, this confirms the mean-convex neighborhood conjecture. Assuming Ilmanen’s multiplicity-one conjecture, we conclude that for embedded two-spheres mean curvature flow through singularities is well-posed.

November 27, 2020 from 15:00 to 16:00 (Montreal/EST time) Zoom meeting

Colloquium presented by **Frances Kirwan (University of Oxford)**

Moduli spaces arise naturally in classification problems in geometry. The study of the moduli spaces of nonsingular complex projective curves (or equivalently of compact Riemann surfaces) goes back to Riemann himself in the nineteenth century. The construction of the moduli spaces of stable curves of fixed genus is one of the classical applications of Mumford's geometric invariant theory (GIT), developed in the 1960s; many other moduli spaces of 'stable' objects can be constructed using GIT and in other ways. A projective curve is stable if it has only very mild singularities (nodes) and its automorphism group is finite; similarly in other contexts stable objects are usually better behaved than unstable ones.

The aim of this talk is to explain how recent methods from a version of GIT for non-reductive group actions can help us to classify singular curves in such a way that we can construct moduli spaces of unstable curves (of fixed type). More generally our aim is to use suitable 'stability conditions' to stratify other moduli stacks into locally closed strata with coarse moduli spaces. The talk is based on joint work with Gergely Berczi, Vicky Hoskins and Joshua Jackson.

November 20, 2020 from 15:00 to 16:00 (Montreal/EST time) Zoom meeting

Colloquium presented by **Wieslawa Niziol (CNRS, Sorbonne University)**

**Chaire Aisenstadt Chair Conference**

**Thematic Semester: Number Theory - Cohomology in Arithmetic**

p-adic Hodge Theory is one of the most powerful tools in modern Arithmetic Geometry. In this talk, I will review p-adic Hodge Theory of algebraic varieties, present current developments in p-adic Hodge Theory of analytic varieties, and discuss some of its applications to problems in Number Theory.

As part of the **Thematic Semester**, Wieslawa Niziola will give a series of four lectures including this one (Hodge Theory of p-adic varieties). The following three will take place from 9:30 a.m. to 10:30 a.m. on Monday November 30, Tuesday December 1st and Wednesday December 2, 2020. Info and registration: http://www.crm.umontreal.ca/2020/Niziol20/index_e.php

November 13, 2020 from 15:30 to 16:30 (Montreal/EST time) Zoom meeting

Colloquium presented by **Tamara Broderick (Massachusetts Institute of Technology, USA)**

The error or variability of statistical and machine learning algorithms is often assessed by repeatedly re-fitting a model with different weighted versions of the observed data. The ubiquitous tools of cross-validation (CV) and the bootstrap are examples of this technique. These methods are powerful in large part due to their model agnosticism but can be slow to run on modern, large data sets due to the need to repeatedly re-fit the model. We use a linear approximation to the dependence of the fitting procedure on the weights, producing results that can be faster than repeated re-fitting by orders of magnitude. This linear approximation is sometimes known as the "infinitesimal jackknife" (IJ) in the statistics literature, where it has mostly been used as a theoretical tool to prove asymptotic results. We provide explicit finite-sample error bounds for the infinitesimal jackknife in terms of a small number of simple, verifiable assumptions. Without further modification, though, we note that the IJ deteriorates in accuracy in high dimensions and incurs a running time roughly cubic in dimension. We additionally show, then, how dimensionality reduction can be used to successfully run the IJ in high dimensions when data is sparse or low rank. Simulated and real-data experiments support our theory.

October 16, 2020 from 15:00 to 16:00 (Montreal/EST time) Zoom meeting

Colloquium presented by **Nicolas Bergeron (École normale supérieure (Paris), France)**

**Chaire Aisenstadt Chair Conference**

**Thematic Semester: Number Theory - Cohomology in Arithmetic**

In his fantastic book “Elliptic functions according to Eisenstein and Kronecker”, Weil writes:

“As Eisenstein shows, his method for constructing elliptic functions applies beautifully to the simpler case of the trigonometric functions. Moreover, this case provides […] the simplest proofs for a series of results, originally discovered by Euler.”

The results Weil alludes to are relations between product of trigonometric functions. I will first explain how these relations are quite surprisingly governed by relations between modular symbols (whose elementary theory I will sketch). I will then show how this story fits into a wider picture that relates the topological world of group homology of some linear groups to the algebraic world of trigonometric and elliptic functions. To conclude I will briefly describe a number theoretical application.

This is based on a work-in-progress with Pierre Charollois, Luis Garcia and Akshay Venkatesh.

October 9, 2020 from 15:00 to 16:00 (Montreal/EST time) Zoom meeting

Colloquium presented by **Phillip Griffiths (Institute for Advanced Study, Princeton, USA)**

The theory of moduli is an important and active area in algebraic geometry. For varieties of general type the existence of a moduli space** **with a canonical completion has been proved by Kollar/Shepard-Barron/Alexeev. Aside from the classical case of algebraic curves, very little is known about the structure of , especially it’s boundary

In this talk, we will discuss some aspects of this topic with emphasis on I-surfaces, which provide one of the first examples where the theory has been worked out in some detail. Particular notice will me made of how the extension data in the limiting mixed Hodge structures that arise from singular surfaces on the boundary of moduli may be used to guide the desingularization of that boundary.

October 2, 2020 from 15:30 to 16:30 (Montreal/EST time) Zoom meeting

Colloquium presented by **Paul McNicholas (McMaster University, Canada)**

Data science is discussed along with some historical perspective. Selected problems in classification are considered, either via specific datasets or general problem types. In each case, the problem is introduced before one or more potential solutions are discussed and applied. The problems discussed include data with outliers, longitudinal data, and three-way data. The proposed approaches are generally mixture model-based.

Organizers:

Erica E. M. Moodie (erica.moodie@mcgill.ca)

Yogendra P. Chaubey (yogen.chaubey@concordia.ca)

September 11, 2020 from 16:00 to 17:00 (Montreal/EST time) Zoom meeting

Colloquium presented by **Stefan Wager (Stanford University, USA)**

Given advances in machine learning over the past decades, it is now possible to accurately solve difficult non-parametric prediction problems in a way that is routine and reproducible. In this talk, I’ll discuss how machine learning tools can be rigorously integrated into observational study analyses, and how they interact with classical statistical ideas around randomization, semiparametric modeling, double robustness, etc. I’ll also survey some recent advances in methods for treatment heterogeneity. When deployed carefully, machine learning enables us to develop causal estimators that reflect an observational study design more closely than basic linear regression based methods.

June 19, 2020 from 16:00 to 17:00 (Montreal/EST time) Zoom meeting

Colloquium presented by **Morgan Craig (Université de Montréal)**

COVID-19 is typically characterized by a range of respiratory symptoms that, in severe cases, progress to acute respiratory distress syndrome (ARDS). These symptoms are also frequently accompanied by a range of inflammatory indications, particularly hyper-reactive and dysregulated inflammatory responses in the form of cytokine storms and severe immunopathology. Much remains to be uncovered about the mechanisms that lead to disparate outcomes in COVID-19. Here, quantitative approaches, especially mechanistic mathematical models, can be leveraged to improve our understanding of the immune response to SARS-CoV-2 infection.

Building upon our prior work modelling the production of innate immune cell subsets and the viral dynamics of HIV and oncolytic viruses, we are developing a quantitative framework to interrogate open questions about the innate and adaptive immune reaction in COVID-19. In this talk, I will outline our recent work modelling SARS-CoV-2 viral dynamics and the ensuing immune response at both the tissue and systemic levels. A portion of this work is done as part of an international and multidisciplinary coalition working to establish a comprehensive tissue simulator (physicell.org/covid19 [1]), which I will also discuss in more detail.

April 17, 2020 from 16:00 to 17:00 (Montreal/EST time) Zoom meeting

Colloquium presented by **Lai-Sang Young (New York University Courant)**

The terms "observable events" and "typical trajectories" in the title should really be between quotation marks, because what is typical and/or observable is a matter of interpretation. For dynamical systems on finite dimensional spaces, one often equates observable events with positive Lebesgue measure sets, and invariant distributions that reflect the large-time behaviors of positive Lebesgue measure sets of initial conditions (such as Liouville measure for Hamiltonian systems) are considered to be especially important. I will begin by introducing these concepts for general dynamical systems -- including those with attractors -- describing a simple dynamical picture that one might hope to be true. This picture does not always hold, unfortunately, but a small amount of random noise will bring it about. In the second part of my talk I will consider infinite dimensional systems such as semi-flows arising from dissipative evolutionary PDEs. I will discuss the extent to which the ideas above can be generalized to infinite dimensions, and propose a notion of "typical solutions".

May 16, 2019 from 16:00 to 18:00 (Montreal/EST time) On location

Colloquium presented by **Jungkay A. Chen (National Taiwan University)**

One of the main themes of algebraic geometry is to classify algebraic varieties and to study various geometric properties of each of the interesting classes. Classical theories of curves and surfaces give a beautiful framework of classification theory. Recent developments provide more details in the case of dimension three. We are going to introduce the three-dimensional story and share some expectations for even higher dimensions.

**Address**

May 10, 2019 from 16:00 to 18:00 (Montreal/EST time) On location

Colloquium presented by **Amanda Folsom (Amherst College)**

Quantum modular forms were defined in 2010 by Zagier; they are somewhat analogous to ordinary modular forms, but they are defined on the rational numbers as opposed to the upper half complex plane, and have modified transformation properties. In 2016, Bringmann and the author defined the notion of a quantum Jacobi form, naturally marrying the concept of a quantum modular form with that of a Jacobi form (the theory of which was developed by Eichler and Zagier in the 1980s). We will discuss these intertwined topics, emphasizing recent developments and applications. In particular, we will discuss applications to combinatorics, topology (torus knots), and representation theory (VOAs).

**Address**

May 3, 2019 from 16:00 to 18:00 (Montreal/EST time) On location

Colloquium presented by **Lenya Ryzhik (Stanford University)**

The stochastic heat equation and the KPZ equation appear as the macroscopic limits for a large class of probabilistic models, and the study of KPZ, in particular, led to many fascinating developments in probability over the last decade or so, from the regularity structures to integrable probability. We will discuss a small group of recent results on these equations in simple settings, of the PDE flavour, that fall in line with what one may call naive expectations by an applied mathematician.

**Address**

April 26, 2019 from 16:00 to 18:00 (Montreal/EST time) On location

Colloquium presented by **Alan W. Reid (Rice University)**

If G is a finitely generated group, let C(G) denote the set of finite quotients of G. This talk will survey work on the question of to what extent C(G) determines G up to isomorphism, culminating in a discussion of examples of Fuchsian and Kleinian groups that are determined by C(G) (amongst finitely generated residually finite groups).

**Address**

April 12, 2019 from 16:00 to 18:00 (Montreal/EST time) On location

Colloquium presented by **Nicolas Bergeron (École normale supérieure (Paris), France)**

Torus bundles over the circle are among the simplest and cutest examples of 3- dimensional manifolds. After presenting some of these examples, using in particular animations realized by Jos Leys, I will consider periodic orbits in these fiber bundles over the circle. We will see that their linking numbers --- that are rational numbers by definition --- can be computed as certain special values of Hecke L-functions. Properly generalized this viewpoint makes it possible to give new topological proof of now classical rationality or integrality theorems of Klingen-Siegel and Deligne-Ribet. It also leads to interesting new "arithmetic lifts" that I will briefly explain. All this is extracted from an on going joint work with Pierre Charollois, Luis Garcia and Akshay Venkatesh.

**Address**

March 29, 2019 from 16:00 to 18:00 (Montreal/EST time) On location

Colloquium presented by **Minhyong Kim (University of Oxford)**

Principal bundles and their moduli have been important in various aspects of physics and geometry for many decades. It is perhaps not so well-known that a substantial portion of the original motivation for studying them came from number theory, namely the study of Diophantine equations. I will describe a bit of this history and some recent developments.

**Address**

March 22, 2019 from 16:00 to 18:00 (Montreal/EST time) On location

Colloquium presented by **Emmy Murphy (Northwestern University)**

We discuss a number of h-principle phenomena which were recently discovered in the field of contact and symplectic geometry. In generality, an h-principle is a method for constructing global solutions to underdetermined PDEs on manifolds by systematically localizing boundary conditions. In symplectic and contact geometry, these strategies typically are well suited for general constructions and partial classifications. Some of the results we discuss are the characterization of smooth manifolds admitting contact structures, high dimensional overtwistedness, the symplectic classification of flexibile Stein manifolds, and the construction of exotic Lagrangians in C^n.

**Address**

March 19, 2019 from 14:30 to 16:30 (Montreal/EST time) On location

Colloquium presented by **Andrew Marks (UCLA)**

In 1925, Tarski posed the problem of whether a disc in R^2 can be partitioned into finitely many pieces which can be rearranged by isometries to form a square of the same area. Unlike the Banach-Tarski paradox in R^3, it can be shown that two Lebesgue measurable sets in R^2 cannot be equidecomposed by isometries unless they have the same measure. Hence, the disk and square must necessarily be of the same area. In 1990, Laczkovich showed that Tarski’s circle squaring problem has a positive answer using the axiom of choice. We give a completely constructive solution to the problem and describe an explicit (Borel) way to equidecompose a circle and a square. This answers a question of Wagon. Our proof has three main ingredients. The first is work of Laczkovich in Diophantine approximation. The second is recent progress in a research program in descriptive set theory to understand how the complexity of a countable group is related to the complexity of the equivalence relations generated by its Borel actions. The third ingredient is ideas coming from the study of flows in networks. This is joint work with Spencer Unger.

**Address**

March 15, 2019 from 16:00 to 18:00 (Montreal/EST time) On location

Colloquium presented by **Shmuel Weinberger (University of Chicago)**

Persistent homology is a very simple idea that was initially introduced as a way of understanding the underlying structure of an object from, perhaps noisy, samples of the object, and has been used as a tool in biology, material sciences, mapping and elsewhere. I will try to explain some of this, but perhaps also some more mathematical applications within geometric group theory. Then I'd like to pivot and study the part that traditionally has been thrown away, and show that this piece is relevant to approximation theory (a la Chebyshev), closed geodesics (a la Gromov), and to problems of quantitative topology (joint work with Ferry, Chambers, Dotter, and Manin).

**Address**

November 2, 2018 from 16:00 to 17:00 (Montreal/EST time)

A strong form of the P ≠ NP conjecture holds that no algorithm faster than n^{O(k)} solves the k-clique problem with high probability when the input is an Erdös–Rényi random graph with an appropriate edge density. Toward this conjecture, I will describe a line of work lower-bounding the average-case complexity of k-clique (and other subgraph isomorphism problems) in weak models of computation: namely, restricted classes of booleancircuits and formulas. Along the way I will discuss some of the history and current frontiers in Circuit Complexity. Joint work with Ken-ichi Kawarabayashi, Yuan Li and Alexander Razborov.

**Address**

September 28, 2018 from 16:00 to 17:00 (Montreal/EST time) On location

Colloquium presented by **Hans-Otto Walther (Universität Giessen)**

Simple-looking autonomous delay differential equations with a real function and single time lag can generate complicated (chaotic) solution behaviour, depending on the shape of . The same could be shown for equations with a variable, state-dependent delay , even for the linear case with . Here the argument of the {\it delay functional} is the history of the solution between and t defined as the function given by . So the delay alone may be responsible for complicated solution behaviour. In both cases the complicated behaviour which could be established occurs in a thin dust-like invariant subset of the infinite-dimensional Banach space or manifold of functions on which the delay equation defines a nice semiflow. The lecture presents a result which grew out of an attempt to obtain complicated motion on a larger set with non-empty interior, as certain numerical experiments seem to suggest. For some we construct a delay functional , an infinite-dimensional subset of the space , so that the equation has a solution whose {\it short segments} , , are dense in the space . This implies a new kind of complicated behaviour of the flowline . Reference: H. O. Walther, {\em A delay differential equation with a solution whose shortened segments are dense}.\\ J. Dynamics Dif. Eqs., to appear.

**Address**

September 21, 2018 from 16:00 to 17:00 (Montreal/EST time) On location

Colloquium presented by **Ezra Miller (Duke University)**

This talk introduces an algebraic framework to encode, compute, and analyze topological summaries of data. The main motivating problem, from evolutionary biology, involves statistics on a dataset comprising images of fruit fly wing veins, which amount to embedded planar graphs with varying combinatorics. Additional motivation comes from statistics more generally, the goal being to summarize unknown probability distributions from samples. The algebraic structures for topological summaries take their cue from graded polynomial rings and their modules, but the theory is complicated by the passage from integer exponent vectors to real exponent vectors. The key to making the structures practical for data science applications is a finiteness condition that encodes topological tameness -- which occurs in all modules arising from data -- robustly, in equivalent combinatorial and homological algebraic ways. Out of the tameness condition surprisingly falls much of ordinary commutative algebra, including syzygy theorems and primary decomposition.

**Address**

May 4, 2018 from 16:00 to 18:00 (Montreal/EST time) On location

Colloquium presented by **Emmanuel Hebey (Université de Cergy-Pontoise)**

We intend to give a general talk about Klein-Gordon-Maxwell-Proca systems which we aim to be accessible to a broad audience. We will insist on the Proca contribution and then discuss the kind of results one can prove in the electromagneto static case of the equations.

**Address**

April 13, 2018 from 16:00 to 18:00 (Montreal/EST time) On location

Colloquium presented by **Eva Bayer-Fluckiger (École Polytechnique Fédérale de Lausanne)**

One of the classical tools of number theory is the socalled localglobal principle, or Hasse principle, going back to Hasse's work in the 1920's. His first results concern quadratic forms, and norms of number fields. Over the years, many positive and negative results were proved, and there is now a huge number of results in this topic. This talk will present some old and new results, in particular in the continuation of Hasse's cyclic norm theorem. These have been obtained jointly with Parimala and Tingyu Lee.

**Address**

February 23, 2018 from 16:00 to 18:00 (Montreal/EST time) On location

Colloquium presented by **Sabin Cautis (University of British Columbia)**

The affine Grassmannian, though a somewhat esoteric looking object at first sight, is a fundamental algebrogeometric construction lying at the heart of a series of ideas connecting number theory (and the Langlands program) to geometric representation theory, low dimensional topology and mathematical physics. Historically it is popular to study the category of constructible perverse sheaves on the affine Grassmannian. This leads to the *constructible* Satake category and the celebrated (geometric) Satake equivalence. More recently it has become apparent that it makes sense to also study the category of perverse *coherent* sheaves (the coherent Satake category). Motivated by certain ideas in mathematical physics this category is conjecturally governed by a cluster algebra structure. We will illustrate the geometry of the affine Grassmannian in an elementary way, discuss what we mean by a cluster algebra structure and then describe a solution to this conjecture in the case of general linear groups.

**Address**

February 16, 2018 from 16:00 to 18:00 (Montreal/EST time) On location

Colloquium presented by **Alexandre Turbiner (UNAM)**

The potential of the n-body problem, both classical and quantum, depends only on the relative (mutual) distances between bodies. By generalized Euler coordinates we mean relative distances and angles. Their advantage over Jacobi coordinates is emphasized. The NEW IDEA is to study trajectories in both classical, and eigenstates in quantum systems which depends on relative distances ALONE. We show how this study is equivalent to the study of (i) the motion of a particle (quantum or classical) in curved space of dimension n(n-1)/2 or the study of (ii) the Euler-Arnold (quantum or classical) - sl(n(n-1)/2, R) algebra top. The curved space of (i) has a number of remarkable properties. In the 3body case the {\it deQuantization} of quantum Hamiltonian leads to a classical Hamiltonian which solves a ~250-years old problem posed by Lagrange on 3-body planar motion.

**Address**

February 16, 2018 from 15:30 to 17:30 (Montreal/EST time) On location

Colloquium presented by **Xiao-Li Meng (Harvard University)**

For over a century now, we statisticians have successfully convinced ourselves and almost everyone else, that in statistical inference the size of the population N can be ignored, especially when it is large. Instead, we focused on the size of the sample, n, the key driving force for both the Law of Large Numbers and the Central Limit Theorem. We were thus taught that the statistical error (standard error) goes down with n typically at the rate of 1/√n. However, all these rely on the presumption that our data have perfect quality, in the sense of being equivalent to a probabilistic sample. A largely overlooked statistical identity, a potential counterpart to the Euler identity in mathematics, reveals a Law of Large Populations (LLP), a law that we should be all afraid of. That is, once we lose control over data quality, the systematic error (bias) in the usual estimators, relative to the benchmarking standard error from simple random sampling, goes up with N at the rate of √N. The coefficient in front of √N can be viewed as a data defect index, which is the simple Pearson correlation between the reporting/recording indicator and the value reported/recorded. Because of the multiplier√N, a seemingly tiny correlation, say, 0.005, can have detrimental effect on the quality of inference. Without understanding of this LLP, “big data” can do more harm than good because of the drastically inflated precision assessment hence a gross overconfidence, setting us up to be caught by surprise when the reality unfolds, as we all experienced during the 2016 US presidential election. Data from Cooperative Congressional Election Study (CCES, conducted by Stephen Ansolabehere, Douglas River and others, and analyzed by Shiro Kuriwaki), are used to estimate the data defect index for the 2016 US election, with the aim to gain a clearer vision for the 2020 election and beyond.

**Address**

February 9, 2018 from 16:00 to 18:00 (Montreal/EST time) On location

Colloquium presented by **Egor Shelukhin (Université de Montréal, Canada)**

In order to resolve Vladimir Arnol'd's famous conjecture from the 1960's, giving lower bounds on the number of fixed points of Hamiltonian diffeomorphisms of a symplectic manifold, Andreas Floer has associated in the late 1980's a homology theory to the Hamiltonian action functional on the loop space of the manifold. It was known for a long time that this homology theory can be filtered by the values of the action functional, yielding information about metric invariants in symplectic topology (Hofer's metric, for example). We discuss a recent marriage between the filtered version of Floer theory and persistent homology, a new field of mathematics that has its origins in data analysis, providing examples of new ensuing results.

**Address**

January 12, 2018 from 16:00 to 18:00 (Montreal/EST time) On location

Colloquium presented by **Semyon Dyatlov (UC Berkeley / MIT)**

Where do eigenfunctions of the Laplacian concentrate as eigenvalues go to infinity? Do they equidistribute or do they concentrate in an uneven way? It turns out that the answer depends on the nature of the geodesic flow. I will discuss various results in the case when the flow is chaotic: the Quantum Ergodicity theorem of Shnirelman, Colin de Verdière, and Zelditch, the Quantum Unique Ergodicity conjecture of Rudnick-Sarnak, the progress on it by Lindenstrauss and Soundararajan, and the entropy bounds of Anantharaman-Nonnenmacher. I will conclude with a recent lower bound on the mass of eigenfunctions obtained with Jin. It relies on a new tool called "fractal uncertainty principle" developed in the works with Bourgain and Zahl.

**Address**

December 8, 2017 from 16:00 to 16:00 (Montreal/EST time) On location

Colloquium presented by **James Maynard (University of Oxford)**

Many famous open questions about primes can be interpreted as questions about the digits of primes in a given base. We will talk about recent work showing there are infinitely many primes with no 7 in their decimal expansion. (And similarly with 7 replaced by any other digit.) This shows the existence of primes in a 'thin' set of numbers (sets which contain at most X^{1c} elements less than X) which is typically very difficult. The proof relies on a fun mixture of tools including Fourier analysis, Markov chains, Diophantine approximation, combinatorial geometry as well as tools from analytic number theory

**Address**

November 24, 2017 from 15:30 to 17:30 (Montreal/EST time) On location

Colloquium presented by **David R. Bellhouse (Western University, London, Ontario)**

As Canada celebrates its 150th anniversary, it may be good to reflect on the past and future of data analysis and statistics in this country. In this talk, I will review the Victorian Statistics Movement and its effect in Canada, data analysis by a Montréal physician in the 1850s, a controversy over data analysis in the 1850s and 60s centred in Montréal, John A. MacDonald’s use of statistics, the Canadian insurance industry and the use of statistics, the beginning of mathematical statistics in Canada, the Fisherian revolution, the influence of Fisher, Neyman and Pearson, the computer revolution, and the emergence of data science.

**Address**

November 24, 2017 from 15:30 to 17:30 (Montreal/EST time) On location

Over the last decades, there was much progress in understanding 2D lattice models of critical phenomena. It started with several theories, developed by physicists. Most notably, Conformal Field Theory led to spectacular predictions for 2D lattice models: e.g., critical percolation cluster a.s. has Hausdorff dimension $91/48$, while the number of selfavoiding length $N$ walks on the hexagonal lattice grows like $(\sqrt{2+\sqrt{2}})^N N^{11/32}$. While the algebraic framework of CFT is rather solid, rigorous arguments relating it to lattice models were lacking. More recently, mathematical approaches were developed, allowing not only for rigorous proofs of many such results, but also for new physical intuition. We will discuss some of the applications of complex analysis to the study of 2D lattice models.

**Address**

November 17, 2017 from 16:00 to 18:00 (Montreal/EST time) On location

Colloquium presented by **Jun-Cheng Wei (UBC)**

Classifying solutions to nonlinear partial differential equations are fundamental research in PDEs. In this talk, I will report recent progress made in classifying some elementary PDEs, starting with the De Giorgi Conjecture (1978). I will discuss the classification of global minimizers and finite Morse index solutions, relation with minimal surfaces and Toda integrable systems, as well as recent exciting developments in fractional De Giorgi Conjecture.

**Address**

October 27, 2017 from 16:00 to 18:00 (Montreal/EST time) On location

Colloquium presented by **Justin Solomon (M)**

Algorithms for analyzing 3D surfaces find application in diverse fields from computer animation to medical imaging, manufacturing, and robotics. Reflecting a bias dating back to the early development of differential geometry, a disproportionate fraction of these algorithms focuses on discovering intrinsic shape properties, or those measurable along a surface without considering the surrounding space. This talk will summarize techniques to overcome this bias by developing a geometry processing pipeline that treats intrinsic and extrinsic geometry democratically. We describe theoreticallyjustified, stable algorithms that can characterize extrinsic shape from surface representations. In particular, we will show two strategies for computational extrinsic geometry. In our first approach, we will show how the discrete LaplaceBeltrami operator of a triangulated surface accompanied with the same operator for its offset determines the surface embedding up to rigid motion. In the second, we will treat a surface as the boundary of a volume rather than as a thin shell, using the Steklov (DirichlettoNeumann) eigenproblem as the basis for developing volumetric spectral shape analysis algorithms without discretizing the interior.

**Address**

October 13, 2017 from 16:00 to 18:00 (Montreal/EST time) On location

Colloquium presented by **Avi Soffer (Rutgers University)**

I will review the problem of Global existence for dispersive equations, in particular, supercritical equations. These equations who play a fundamental role in science, have been , and remain a major challenge in the field of Partial Differential Equations. They come in various forms, derived from Geometry, General Relativity, Fluid Dynamics, Field Theory. I present a new approach to classify the asymptotic behavior of wave equations, supercritical and others, and construct global solutions with large initial data. I will then describe current extensions to Nonlinear Schroedinger Equations.

**Address**

September 29, 2017 from 16:00 to 18:00 (Montreal/EST time) On location

Colloquium presented by **John H. Conway (Princeton University)**

The “first field” is obtained by making the entries in its addition and multiplication tables be the smallest possibilities. It is really an interesting field that contains the integers, but with new addition and multiplication tables. For example, 2 x 2 = 3, 5 x 7 = 13, ... It extends to the infinite ordinals and the first infinite ordinal is the cube root of 2!

**Address**

September 15, 2017 from 16:00 to 18:00 (Montreal/EST time) On location

In this talk, we will first review the classic Weyl's embedding problem and its application in quasilocal mass. We will then discuss some recent progress on Weyl's embedding problem in general Riemannian manifold. Assuming isometric embedding into Schwarzschild manifold, we will further establish a quasilocal type inequality. This talk is based on works joint with Pengfei Guan and Pengzi Miao.

**Address**

May 5, 2017 from 16:00 to 16:00 (Montreal/EST time) On location

Colloquium presented by **Gerard Freixas (Institut de Mathématiques de Jussieu)**

Arakelov geometry is a modern formalism that extends in various directions the geometry of numbers founded by Minkowski in the nineteenth century. The objects of study are arithmetic varieties, namely complex varieties that can be defined by polynomial equations with integer coefficients. The theory exploits the interplay between algebraic geometry and number theory and complex analysis and differential geometry. Recently, the formalism found beautiful and important applications to the socalled Kudla programme and the Colmez conjecture. In the talk, I will first introduce elementary facts in Minkowski's geometry of numbers. This will provide a motivation for the sequel, where I will give my own view of Arakelov geometry, by focusing on toy (but nontrivial) examples of one of the central theorems in the theory, the arithmetic RiemannRoch theorem mainly due to Bismut, Gillet and Soulé, and generalizations. I hope there will be ingredients to satisfy different tastes, for instance modular forms (arithmetic aspect), analytic torsion (analytic aspect) and Selberg zeta functions (arithmetic, analytic and dynamic aspects).

**Address**

April 21, 2017 from 16:00 to 16:00 (Montreal/EST time) On location

Colloquium presented by **Aaron Naber (Northwestern University)**

In this talk we give an introduction to the analysis of the YangMills equation in higher dimensions. In particular, when studying sequences of solutions we will study the manner in which blow up can occur, and how this blow up may be understood through the classical notions of the defect measure and bubbles. The energy identity is an explicit conjectural relationship, known to be true in dimension four, relating the energy density of the defect measure at a point to the bubbles which occur at that point, and we will give a brief overview of the recent proof of this result for general stationary Yang Mills in higher dimensions. The work is joint with Daniele Valtorta.

**Address**

March 31, 2017 from 16:00 to 18:00 (Montreal/EST time) On location

Colloquium presented by **Tatiana Toro (University of Washington)**

In these lecture we will discuss the relationship between the boundary regularity of the solutions to elliptic second order divergence form partial differential equations and the geometry of the boundary of the domain where they are defined. While in the smooth setting tools from classical PDEs are used to address this question, in the nonsmooth setting techniques from harmonic analysis and geometric measure theory are needed to tackle the problem. The goal is to present an overview of the recent developments in this very active area of research.

**Address**

March 17, 2017 from 15:30 to 17:30 (Montreal/EST time) On location

Colloquium presented by **Sayan Mukherjee (Duke University)**

We consider the asymptotic consistency of maximum likelihood parameter estimation for dynamical systems observed with noise. Under suitable conditions on the dynamical systems and the observations, we show that maximum likelihood parameter estimation is consistent. Furthermore, we show how some wellstudied properties of dynamical systems imply the general statistical properties related to maximum likelihood estimation. Finally, we exhibit classical families of dynamical systems for which maximum likelihood estimation is consistent. Examples include shifts of finite type with Gibbs measures and Axiom A attractors with SRB measures. We also relate Bayesian inference to the thermodynamic formalism in tracking dynamical systems.

**Address**

March 10, 2017 from 16:00 to 18:00 (Montreal/EST time) On location

Colloquium presented by **Louigi Addario-Berry (Université McGill)**

One of the most dynamic areas of probability theory is the study of the behaviour of discrete optimization problems on random inputs. My talk will focus on the probabilistic analysis of one of the first and foundational combinatorial optimization problems: the minimum spanning tree problem. The structure of a random minimum spanning tree (MST) of a graph G turns out to be intimately linked to the behaviour of critical and nearcritical percolation on G. I will describe this connection, and present some results on the structure, scaling limits, and volume growth of random MSTs. It turns out that, on highdimensional graphs, random minimum spanning trees are expected to be threedimensional when viewed intrinsically, and sixdimensional when viewed as embedded objects. Based on joint works with Nicolas Broutin, Christina Goldschmidt, Simon Griffiths, Ross Kang, Gregory Miermont, Bruce Reed, Sanchayan Sen.

**Address**

February 24, 2017 from 16:00 to 18:00 (Montreal/EST time) On location

Colloquium presented by **Frithjof Lutscher (Université d'Ottawa)**

The globally observed phenomenon of the spread of invasive biological species with all its sometimes detrimental effects on native ecosystems has spurred intense mathematical research and modelling efforts into corresponding phenomena of spreading speeds and travelling waves. The standard modelling framework for such processes is based on reaction diffusion equations, but several aspects of an invasion can only be appropriately described by a discretetime analogues, called integrodifference equations. The theory of spreading speeds and travelling waves in such integrodifference equations is well established for the "monostable" case, i.e. when the nonspatial dynamics show a globally stable positive steady state. When the positive state of the nonspatial dynamics is not stable, as is the case with the famous discrete logistic equation, it is unclear how the corresponding spatial spread profile evolves and at what speed. Previous simulations seemed to reveal a travelling profile in the form of a twocycle, with or without spatial oscillations. The existence of a travelling wave solution has been proven, but its shape and stability remain unclear. In this talk, I will show simulations that suggest that there are several travelling profiles at different speeds. I will establish corresponding generalizations of the concept of a spreading speed and prove the existence of such speeds and travelling waves in the second iterate operator. I conjecture that rather than a travelling twocycle for the nextgeneration operator, one observes a pair of stacked fronts for the seconditerate operator. I will relate the observations to the phenomenon of dynamic stabilization.

**Address**

February 10, 2017 from 16:00 to 18:00 (Montreal/EST time) On location

Colloquium presented by **Mark Powell (UQAM)**

I will introduce the knot concordance group, give a survey of our current understanding of it and discuss some relationships with the topology of 4-manifolds.

**Address**

January 20, 2017 from 16:00 to 18:00 (Montreal/EST time) On location

Colloquium presented by **Christopher Skinner (Princeton University)**

This colloquium talk will begin with an introduction to the Birch-Swinnerton-Dyer conjecture for elliptic curves -- just curves defined by the equations y^2=x^3+Ax+B -- and then describe recent advances that allow us to prove that lots of elliptic curves have rank zero or one.

**Address**

December 2, 2016 from 16:00 to 18:00 (Montreal/EST time) On location

As is well-known, two of the basic types of linear partial differential equations (PDEs) are hyperbolic PDEs and elliptic PDEs, following the classification for linear PDEs first proposed by Jacques Hadamard in the 1920s; and linear theories of PDEs of these two types have been well established, respectively. On the other hand, many nonlinear PDEs arising in mechanics, geometry, and other areas naturally are of mixed elliptichyperbolic type. The solution of some longstanding fundamental problems in these areas greatly requires a deep understanding of such nonlinear PDEs of mixed type. Important examples include shock reflection-diffraction problems in fluid mechanics (the Euler equations) and isometric embedding problems in differential geometry (the Gauss-CodazziRicci equations), among many others. In this talk we will present natural connections of nonlinear PDEs of mixed elliptic-hyperbolic type with these longstanding problems and will then discuss some recent developments in the analysis of these nonlinear PDEs through the examples with emphasis on developing and identifying mathematical approaches, ideas, and techniques for dealing with the mixed-type problems. Further trends, perspectives, and open problems in this direction will also be addressed.

**Address**

December 1, 2016 from 15:30 to 17:30 (Montreal/EST time) On location

Colloquium presented by **Richard Samworth (University of Cambridge)**

Changepoints are a very common feature of Big Data that arrive in the form of a data stream. We study highdimensional time series in which, at certain time points, the mean structure changes in a sparse subset of the coordinates. The challenge is to borrow strength across the coordinates in order to detect smaller changes than could be observed in any individual component series. We propose a twostage procedure called 'inspect' for estimation of the changepoints: first, we argue that a good projection direction can be obtained as the leading left singular vector of the matrix that solves a convex optimisation problem derived from the CUSUM transformation of the time series. We then apply an existing univariate changepoint detection algorithm to the projected series. Our theory provides strong guarantees on both the number of estimated changepoints and the rates of convergence of their locations, and our numerical studies validate its highly competitive empirical performance for a wide range of data generating mechanisms.

**Address**

November 26, 2016 from 16:00 to 18:00 (Montreal/EST time) On location

Colloquium presented by **Maksym Radziwill (Northwestern)**

The Moebius function plays a central role in number theory; both the prime number theorem and the Riemann Hypothesis are naturally formulated in terms of the amount of cancellations one gets when summing the Moebius function. In recent joint work with K. Matomaki the speaker showed that the sum of the Moebius function exhibits cancellations in "almost all intervals'' of increasing length. This goes beyond what was previously known conditionally on the Riemann Hypothesis. The result holds in fact in greater generality. Exploiting this generality one can show that between a fixed number of consecutive squares there is always an integer composed of only "small'' prime factors. This is related to the running time of Lenstra's factoring algorithm. I will also discuss some further developments : the work of Tao on correlations between consecutive values of Chowla, and his application of this result to the resolution of the Erdos discrepancy problem.

**Address**

November 4, 2016 from 16:00 to 18:00 (Montreal/EST time) On location

Colloquium presented by **Philippe G. LeFloch (Université Pierre et Marie Curie, Paris 6)**

will review results on the global evolution of selfgravitating massive matter in the context of Einstein's theory as well as the f(R)theory of gravity. In collaboration with Yue Ma (Xian), I have investigated the global existence problem for the Einstein equations coupled with a KleinGordon equation describing the evolution of a massive scalar field. Our main theorem establishes the global nonlinear stability of Minkowski spacetime upon small perturbations of the metric and the matter field. Recall that the fully geometric proof by Christodoulou and Klainerman in 1993, as well as the proof in wave gauge by Lindblad and Rodnianski in 2010, both apply to vacuum spacetimes and massless fields only. Our new technique of proof, which we refer to as the Hyperboloidal Foliation Method, does not use Minkowski's scaling field and is based on a foliation of the spacetime by asymptotically hyperboloidal spacelike hypersurfaces, on sharp estimates for wave and KleinGordon equations, and on an analysis of the quasinull hyperboloidal structure (as we call it) of the Einstein equations in wave gauge.

**Address**

October 28, 2016 from 15:30 to 17:30 (Montreal/EST time) On location

Colloquium presented by **Jerry Lawless (University of Waterloo)**

Two-phase studies are frequently used when observations on certain variables are expensive or difficult to obtain. One such situation is when a cohort exists for which certain variables have been measured (phase 1 data); then, a subsample of individuals is selected, and additional data are collected on them (phase 2). Efficiency for tests and estimators can be increased by basing the selection of phase 2 individuals on data collected at phase 1. For example, in large cohorts, expensive genomic measurements are often collected at phase 2, with oversampling of persons with “extreme” phenotypic responses. A second example is casecohort or nested casecontrol studies involving times to rare events, where phase 2 oversamples persons who have experienced the event by a certain time. In this talk I will describe twophase studies on failure times, present efficient methods for testing covariate effects. Some extensions to more complex outcomes and areas needing further development will be discussed.

**Address**

October 21, 2016 from 16:00 to 16:00 (Montreal/EST time) On location

Colloquium presented by **Ivan Corwin (Columbia University)**

I will explain how certain integrable structures give rise to meaningful probabilistic systems and methods to analyze them. Asymptotics reveal universal phenomena, such as the KardarParisiZhang universality class. No prior knowledge will be assumed.

**Address**

October 14, 2016 from 16:00 to 18:00 (Montreal/EST time) On location

Colloquium presented by **Jean-Philippe Lessard (Centre de recherches mathématiques)**

Studying and proving existence of solutions of nonlinear dynamical systems using standard analytic techniques is a challenging problem. In particular, this problem is even more challenging for partial differential equations, variational problems or functional delay equations which are naturally defined on infinite dimensional function spaces. The goal of this talk is to present rigorous numerical technique relying on functional analytic and topological tools to prove existence of steady states, time periodic solutions, traveling waves and connecting orbits for the above mentioned dynamical systems. We will spend some time identifying difficulties of the proposed approach as well as time to identify future directions of research.

**Address**