Department of Mathematics and Statistics

Department of Mathematics and Statistics
Department of Mathematics and Statistics
Subscribe to RSS - Seminars

Geometry & Representation - Kaveh Mousavand (Queen's University)

Monday, October 26th, 2020

Time: 4:30p.m.  Place: Zoom

Speaker: Kaveh Mousavand (Queen's University)

Title: A Categorification of Biclosed Sets of Strings.

Abstract:  In joint work with A. Garver and T. McConville, for any gentle algebra of finite representation type we studied a closure operator on the set of strings. Shortly before, Palu, Pilaud, and Plamondon had proved that the collection of all biclosed sets of strings forms a lattice, and moreover, that this lattice is congruence-uniform. Many interesting examples of finite congruence-uniform lattices may be represented as the lattice of torsion classes of an associative algebra. To extend this result, we introduced a generalization-- the lattice of torsion shadows-- and proved that the lattice of biclosed sets of strings is isomorphic to a lattice of torsion shadows.

Furthermore, we introduced the analogous notion of wide shadows to extend wide subcategories, so we could find a new realization for another important lattice theoretical phenomenon. In fact, we showed the shard intersection order of the lattice of biclosed sets is isomorphic to a lattice of wide shadows. Consequently, in our setting, we established a bijection between torsion shadows and wide shadows, analogous to those between functorially finite torsion classes and wide subcategories studied by Ingalls-Thomas and Marks-Stovicek.

Website details here:

Department Colloquium - Emine Yildirim (Queen’s University)

Emine Yildirim (Queen’s University)

Friday, October 23rd, 2020

Time: 2:30 p.m.  Place: Online (via Zoom)

Speaker: Emine Yildirim (Queen’s University)

Title: Graphs and Combinatorics in Representation Theory of Algebras.

Abstract: Representation theorists of finite dimensional algebras often use quivers, also known as directed graphs, and many other combinatorial tools associated with these quivers. This is because we understand the module category of algebras via representation of quivers. On the other hand, we also capture the beautiful combinatorics of cluster algebras via the same representation theory. In this talk, I will outline how this machinery works along with some recent results on Cluster Categories we obtained joint with Charles Paquette using the combinatorics and representations theory of quivers.

Emine Yildirim is a Coleman Research Fellow within the Department of Mathematics and Statistics at Queen’s University. She obtained her Ph.D. in Mathematics from the Universite du Quebec a Montreal in 2018. She is mainly interested in representation theory of algebras, specifically path algebras, incidence algebras, and their representations. She also works on cluster algebras, their categorification and related combinatorics.

Dynamics, Geometry, & Groups - Ilya Gekhtman (Technion - Israel)

Thursday, October 22nd, 2020

Time: 2:00 p.m Place:

Speaker: Ilya Gekhtman (Technion-Israel Institute of Technology)

Title: Martin, Floyd and Bowditch boundaries of relatively hyperbolic groups.

Abstract: Consider a transient random walk on a countable group $G$. The Green distance between two points in the group is defined to be minus the boundary of the probability that a random path starting at the first point ever reaches the second. The Martin compactification of the random walk is a topological space defined to be the horofunction boundary of the Green distance. It is a topological model for the Poisson boundary. The Martin boundary typically heavily depends on the random walk; it is thus exciting when for some large class of random walks, the Martin boundary is equivariantly homeomorphic to some well known geometric boundary of the group. Ancona showed in 1988 that this is the case for finitely supported random walks on hyperbolic groups: the Martin boundary is identified with the Gromov boundary. We generalize Ancona's results to relatively hyperbolic groups: the Martin boundary equivariantly continuously surjects onto the Gromov boundary of any hyperbolic space on which the group acts geometrically finitely (called the Bowditch boundary), and the preimage of any conical limit point is a singleton. When the parabolic subgroups are virtually abelian (e.g. for Kleinian groups) we show that the preimage of a parabolic fixed point is a sphere of appropriate dimension, so the Martin boundary can be identified with a Sierpinski carpet. A major technical tool is a generalization of a deviation inequality due to Ancona saying the Green distance is nearly additive along word geodesics, which has various other applications, including to comparing harmonic and Patterson-Sullivan measures for negatively curved manifolds and to local limit theorems for random walks. We do all this using an intermediate construction called the Floyd metric obtaining by suitably rescaling the Cayley graph and considering the associated completion called the Floyd compactification. We show that for any finitely supported random walk on a finitely generated group, the Martin boundary surjects to the Floyd boundary, which in turn by work of Gerasimov covers the Bowditch boundary of relatively hyperbolic groups. This is based on several joint works with subsets of Dussaule, Gerasimov, Potyagailo, and Yang.

Department Colloquium - Rafael Potrie (U de la Republica-Uruguay)

Rafael Potrie (Universidad de la Republica-Uruguay)

Friday, October 16th, 2020

Time: 2:30 p.m.  Place: Online (via Zoom)

Speaker: Rafael Potrie (Universidad de la Republica-Uruguay)

Title: Anosov flows and the fundamental group.

Abstract: The hairy ball theorem says that a vector field in the sphere must have some singularity. How does the dynamics interact with the topology of the underlying manifold in higher dimensions? We will discuss some instances of this question for dynamics in 3-manifolds, featuring a beautiful result due to Margulis and Plante-Thurston. Time permitting, we will touch upon more recent developments on the interactions between the topology and dynamics in 3 dimensions.

Rafael Potrie is an Associate Professor at the Universidad de la Republica-Uruguay. He obtained his Ph.D. in Mathematics from Universite Paris 13/Universidad de la Republica-Uruguay in 2012. He was an invited speaker of the Dynamical Systems session at the International Congress of Mathematicians (ICM) in 2018. His research mainly concerns the topological classification of partially hyperbolic systems in three-dimensional manifolds and its dynamical consequences. Other interests include smooth dynamics, ergodic theory, discrete subgroups of Lie groups, and the geometry of foliations and laminations.

Dynamics, Geometry, & Groups - Mathew Cordes (ETH Zurich)

Thursday, October 15th, 2020

Time: 2:00 p.m Place:

Speaker: Mathew Cordes (ETH Zurich)

Title: Geometric approximate group theory.

Abstract: An approximate group is a group that is "almost closed" under multiplication. Finite approximate subgroups play a major role in additive combinatorics. Recently Breuillard, Green and Tao have established a structure theorem concerning finite approximate subgroups and used this theory to reprove Gromov’s growth theorem. Infinite approximate groups were studied implicitly long before the formal definition. Approximate subgroups of R^n that are Delone sets can be constructed using "cut-and-project" methods and are models for mathematical quasi-crystals. Recently, Björklund and Hartnick have begun a program investigating infinite approximate lattices in locally compact second countable groups using geometric and measurable structures. In the talk I will introduce infinite approximate groups and their geometric aspects. This is joint work with Hartnick and Tonic.

Geometric Sports Analytics - Nick Cuzoj-Shulman (Hockey Data Analyst)

Tuesday, October 13th, 2020

Time: 2:00-3:30 p.m.  Place: Virtual Zoom Meeting

Speakers: Nick Cuzoj-Shulman (Senior Hockey Data Analyst, Sportlogiq; Master of Management Analytics, Smith School of Business)

Seminar Organizers: Catherine Pfaff & Daniel McBride

Information: Please email Catherine Pfaff for more information and to get the Zoom link.

Department Colloquium - Farouk Nathoo (University of Victoria)

Farouk Nathoo  (University of Victoria)

Friday, October 9th, 2020

Time: 2:30 p.m.  Place: Online (via Zoom)

Speaker: Farouk Nathoo (University of Victoria)

Title: Spatial Statistical Modeling for Neuroimaging Data.

Abstract: I will describe three projects involving the analysis of neuroimaging data and the development of hierarchical spatial Bayesian models for each. In the first, we develop an approach for determining the location and dynamics of brain activity from combined magnetoencephalography and electroencephalography data. The resulting inverse problem is ill-posed and we propose a distributed solution based on a Bayesian spatial finite mixture model that incorporates the Potts model to represent the spatial dependence in an allocation process that partitions the cortical surface into a small number of latent states. In the second project, we consider statistical modelling of functional magnetic resonance imaging (fMRI) data which is challenging in part as the data are both spatially and temporally correlated. Motivated by an event‐related fMRI experiment, we propose a novel hierarchical Bayesian model with automatic selection of the auto‐regressive orders of the noise process that vary spatially over the brain. In the third project, we develop a Bayesian bivariate spatial model for multivariate regression analysis applicable to studies examining the influence of genetic variation on brain structure. Our model is motivated by an imaging genetics study of the Alzheimer's Disease Neuroimaging Initiative, where the objective is to examine the association between images of volumetric and cortical thickness values summarizing the structure of the brain as measured by magnetic resonance imaging (MRI) and a set of 486 SNPs from 33 Alzheimer's Disease (AD) candidate genes obtained from 632 subjects. A bivariate spatial process model is developed to accommodate the correlation structures typically seen in structural brain imaging data and we develop a mean-field variational Bayes algorithm and a Gibbs sampling algorithm to fit the model. We compare the new spatial model to an existing non-spatial model in our motivating application.

Dynamics, Geometry, & Groups - Chris Judge

Thursday, October 8th, 2020

Time: 2:00 p.m Place:

Speaker: Chris Judge (Indiana University Bloomington)

Title: Triangles have no interior hot spots in the long run.

Abstract: It well-known that the temperature distribution of a perfectly insulated, perfectly homogeneous body will tend to a constant temperature distribution as time tends to infinity. Jeff Rauch conjectured in the 1974 that, moreover, all local extrema of the temperature distribution will migrate towards the boundary as time tends to infinity. This is now believed to be true for convex domains. I will discuss joint work with Sugata Mondal that verifies the conjecture for triangles. This resolves Polymath 7.