Seminar

This seminar is intended for an undergraduate audience. If you have an interesting talk that would be suitable for undergrads, feel free to contact the organizers and volunteer as a speaker. We especially appreciate talks by undergrads for undergrads!

Seminar runs Wednesdays 14-15:00 (EST) on Zoom. We start seminar at 14:05, to let people arrive and get settled.

After attending seminar, we ask that participants fill out this feedback form. If you are not able to use Google Forms and would still like to give feedback, please e-mail the organizers.

The prime factorization of the Meeting ID is: $11 \times 59 \times 124877917$. The passcode for the meeting is: seminar. If you have trouble accessing the seminar, please contact the co-organizers: Parker Glynn-Adey and Lisa Jeffery. We will send you a direct link.

If you want to receive regular announcements about Seminar, please sign up for our ListServ. We’re also listed on ResearchSeminars.org which you can add to your Google Calendar.

At the bottom of this page, there FAQ for speakers.

• ResearchSeminars.org
• UTStG Seminars (admin)
• The Seminar ListServ
• Feedback Google Form

Winter 2022 Seminar

March 16: Modelling Mathematics with Knitting and Crochet by Kevin Santos

Abstract:

When studying subjects like geometry and topology, it can be difficult to visualize or understand certain abstract concepts. Being able to hold and manipulate a physical model of a mathematical object can give deeper intuition into its properties. The process of constructing such an object also offers further insight. In this talk, we will investigate how geometric and topological objects can be constructed using the crafts of knitting and crochet, which offer unique advantages in creating models. We will describe how the hyperbolic plane can be modelled with crochet and we will explore how topological surfaces such as the sphere, the torus, and the Klein bottle can be knitted. No prior knowledge in geometry, topology, knitting, or crochet is required.

March 23: Why Geometric Algebra Should be in the Standard Linear Algebra Curriculum by Logan Lim

Abstract:

Geometric algebra is an extension of $\mathbb{R}^n$ that includes as special cases: The complex numbers, quaternions, exterior algebras, dual numbers, split-complex numbers, dual quaternions and more! When applied to multivariable calculus, it generalizes the fundamental theorem of calculus on manifolds to include the divergence theorem, curl theorem, and gradient theorem, and as a result Green’s and Stoke’s theorem, as special cases of a single statement. It also simplifies many geometric operations in computer graphics by eliminating the need for matrices in projections, rotations, and reflections. Though we can only cover the ‘main idea’ of geometric algebra in the allotted time, this talk will be a buffet of ideas you can explore for this fascinating and deceptively simple algebraic object.

Mathematicians hate him!!!! See how Clifford generalized rotations and orthogonal complements in n-dimensions with this one simple trick! $$ \mathbf{uv} = \mathbf{u} \cdot \mathbf{v} + \mathbf{u} \wedge \mathbf{v}. $$

March 30: Seifert Surfaces and Knot Genus by Alex Teeter

Abstract:

Knot Theory, a field of mathematics born from a misguided model for atoms, has since grown to become an important subfield of Topology. Not only does it have much of mathematical interest, but also numerous connections to fields such as graph theory, the study of manifolds, and applications to Biology and Physics. We will analyze the connection between Knot Theory and the Topology of Surfaces. Along the way, we will cover the Euler Characteristic, Genus and the beautiful Classification Theorem of Surfaces. Through this consideration, we develop an algorithm to associate each Knot with a surface, and uncover an important invariant, the Genus of a Knot. This will not only allow us to distinguish between different Knots, but will also be vital in establishing fundamental properties of prime and composite Knots. No prior background in Knot Theory or Topology is assumed.

Fall 2021 Seminar

November 17: Fractional Brownian Motion

Abstract: The classical Brownian motion (or Wiener process) serves as the fundamental object of probability theory with vast theoretical and practical applications in a plethora of fields. In the early 1940’s Kolmogorov sought a natural one parameter extension of the process in aims of modelling turbulence in liquids, the extension consists of retaining the framework of the classical motion by constructing still a continuous centered Gaussian process that retains self similarity (of a now distinct index from that of the classical motion) and stationarity of increments but draws its distinction by parameterizing its specifying covariance structure with what is known as the Hurst index, $H \in (0,1)$, resulting in (for ‘most’ values of $H$) a non-Markov process allowing it to serve as a popular model for dependent phenomena; this extension has since been kept in common parlance as fractional Brownian motion (fBm). This talk aims to provide discussion and (at times demonstration) of the fundamental properties of the fBm as well as investigate sample path properties’ dependence on the Hurst parameter.

November 10: Matrix Analysis with a Focus on Inequalities

Abstract: Matrix inequalities are one of the key components in matrix analysis. They have a wide range of applications in statistics, computer science, economics, and physics.

In this presentation, we focus on some classical matrix inequalities such as the Rayleigh-Ritz quotient, the Courant Fischer theorem, Weyl’s inequality, the interlocking eigenvalue lemma, and the Woodbury matrix identity. In particular, some of these show the importance of the maximum and minimum of eigenvalues and singular values.

Recommend Background: Linear Algebra and Calculus. (In the beginning of the presentation, some definitions and key properties such as positive definiteness will be reviewed.)

November 24: Getting Started: Proving with the Lean Interactive Theorem Prover

Abstract: Have you heard of automated and interactive theorem provers? Did you know that American mathematician Alex Kontorovich predicts that the Lean Interactive Theorem Prover will be so widely used that it will be as necessary as LaTex in doing mathematics? This seminar series will take you on a journey, using Lean to give you insights on how to go about proving, which route to choose, how to check for errors, and how to verify a computation or a proof. Particularly, you will be learning to use Lean by playing a number game. Come join us and have some fun!

Summer 2021 Seminar

We are listed on ResearchSeminars.org as UndergraduateSeminar.

Speakers

August 25: Amalrose Vayalinkal: Virtual Ring Routing

Abstract:

Virtual Ring Routing (VRR) schemes define a routing algorithm for communication between devices by establishing a virtual network overlay given a physical network of $N$ devices. Using graphs to model the physical network, we introduce the algorithm and explore the pros and cons of VRR. Time permitting, we take a closer look at the simpler case where the physical network is also a ring (circle) and discuss future directions. This work is part of an NSERC USRA this summer under the supervision of Professor Almut Burchard.

FRIDAY August 20: Xin Ya Xu and Richard Ye: Hero Rats - Detecting Tuberculosis

Abstract:

Tuberculosis has been the leading infectious killer in the world with an infection rate of approximately 10 million people per year. Slow and inaccurate detection methods contribute heavily to making tuberculosis so deadly. In this talk, we will highlight the efforts of APOPO, a non-profit organization that trains scent-detecting African giant pouched rats, to tackle tuberculosis. We will also talk about our approach to improve APOPO’s scent detection technology by using statistical analysis. We will explore the process of applying clustering techniques, such as hierarchical clustering and K-means clustering, to the problem.

Work by Xin Ya Xu and Richard Ye, supervised by Professor Sohee Kang and Professor Marco Pollanen. This is a project where the data was brought to us by Professor Pollanen who works with APOPO, a firm focused on detecting land mines and tuberculosis mainly in sub-saharan Africa.

August 18: Jiawei Chen, Ran Li and Junru Lin: Rapid Testing in COVID and Modified SIR Model

Abstract:

The traditional PCR test for coronavirus typically takes one day to produce results, which cannot satisfy the great volume of demand for testing. However, the recently developed “Rapid Test” (which detects Covid antigen) can help us obtain reasonably accurate results in about 15 minutes. In this talk, we will use a modified SIR model (a classical ODE system for modelling epidemic) to demonstrate how the dynamics of Covid-19 can be affected under the introduction of this new testing technique. We will then discuss how effective the test needs to be in order to control the pandemic.

August 11: Julie Midroni: Artificial neural networks: The fundamentals

Abstract:

Artificial neural networks (ANNs) are a powerful class of machine learning algorithms that can be used for a variety of purposes, including image classification, function approximation, and natural language processing. This talk will present the mathematical basics of ANNs, and briefly explore different ANN algorithms and their uses.

August 4: Andrew Fallone: The Projected Number of Underreported COVID 19 Cases in Canada

Abstract:

As Canada focuses more on mitigation strategies rather than eradication ones, mathematical modelling could play an important role in preserving lives. The model used in this project is a modified SIR model which retains a conservative calculation for COVID 19 cases while giving an insight into how unreported COVID 19 cases are in Canada with the assumption that we have a limited amount of data. This approach seems to be the most effective due to the uncertainty of COVID 19 progression in Canada. Furthermore, this modified SIR model uses a basic reproduction parameter “to estimate the attack rate, epidemic duration”, and critical points of COVID 19 cases in Canada.

July 28: Rakan Omar: Influence Centrality

Abstract:

In graph theory and network analysis, the notion of centrality refers to assigning nodes in a network an index representing the extent to which each node is central - important to the network, or well positioned in it - based on some mathematical property. There are a variety of measures of centrality, each of which measures ‘centrality’ differently, based on (or resulting in) a different definition of ‘importance’ or ‘prominence’.

I will define an original measure of centrality, a variation of pageRank centrality, which I call ‘influence centrality’, that measures the extent to which a node contributes a relation (what is represented by the arcs) to the graph in a directed weighted graph with a finite number of nodes, which are initially labelled with values. I will discuss some properties, applications, and extensions of influence centrality.

I assume some familiarity with graph theory terminology and linear algebra.

July 21: Kevin Santos: An Introduction to Group Theory through Puzzles

Abstract:

The concept of a group is a powerful tool that we can use to understand the structures of mathematical objects. In this talk, we’ll use well-known puzzles such as Peg Solitaire, the 15 puzzle, and Rubik’s cube to motivate a brief introduction to the concepts of group theory. We’ll explore how groups are related to symmetry and give some examples of groups, such as the Klein 4-group and the permutation groups. We’ll then see how these ideas can be applied to understand the puzzles.

July 14: Albert Lai: Partial Orders and Application to the Semantics of Computer Programs

Abstract:

Partial orders generalize total orders by allowing all four possibilities: $x<y$, $x>y$, $x=y$, or none of the above. (Total orders disallow the fourth possibility.) A familiar example is the subset relation over a family of sets. I will show a computer-science application of partial orders to modelling recursive programs. This will be a glimpse of denotational semantics, the study of describing program behaviour by mapping programs to suitable mathematical structures and partial orders.

July 7: Yuveshen Mooroogen: Shining a rainbow-coloured light on the fundamental theorem of algebra

Abstract:

The graphs of real-valued functions on the real line are subsets of a two-dimensional space. As a result, we can sketch them on a piece of paper.

The graphs of complex-valued functions on the complex plane, however, are subsets of a four-dimensional space. Good luck sketching that on a piece of paper.

In this presentation, I will introduce “domain colouring”, which is a technique used to illustrate functions of the complex numbers.

This talk will be in two parts. In the first part, I will explain how to read and construct domain colouring plots, focusing on a number of simple examples. In the second part, I will discuss the Fundamental Theorem of Algebra (which states that every nonconstant polynomial has a complex root) and explain how to visualise one of its proofs using domain colouring. This part of the talk will draw heavily from D. J. Velleman’s beautiful expository article [1].

Prerequisites: Familiarity with the concept of dimension (for a vector space) and with the complex numbers. (Basic principles only. You should know what the notation $x + iy$ means, and how to convert it to modulus-argument/polar form.) Knowledge of multivariable calculus and complex analysis is not expected.

[1] Velleman, D.J. The Fundamental Theorem of Algebra: A Visual Approach. Math Intelligencer 37, 12–21 (2015). https://doi.org/10.1007/s00283-015-9572-7.

June 23 and 30: David Schrittesser: Your life will be better with infinitesimals

Abstract: When Leibniz, Newton, and others first developed calculus, they used the metaphor of infinitely small, or infinitesimal, quantities to try to justify their methods. Later, infinitesimals were expelled from mathematics and calculus was made rigorous using the familiar notions of limit and epsilon-delta formulations.

But infinitesimals have been making a come back! Using methods from logic, in particular model theory, they have been restored as respected citizens in rigorous mathematical arguments. This approach, called non-standard analysis has been described as ``the analysis of the future’'. And indeed, it sometimes allows us to do miraculous things. A recent case in point is my joint result with D. Roy and H. Duanmu, with which we solve a long-standing open problem in statistics (namely giving a Bayesian interpretation of admissibility).

In this talk I will give an introduction to the non-standard method and describe some applications. (And if I manage to spark your interest, come to the course I will teach about this topic at U of T in this fall!)

June 16: Jesse Maltese: An Introduction to Mathematical Logic

Abstract:

In this talk, I will give an introduction to the various ideas underlying the foundations of math. I will first talk through some of the history of the field. Then, I will introduce basic concepts in logic including the definition of truth, a model, and notions like completeness and consistency. Then, I will state two important theorems in the field: Gödel’s Completeness Theorem, and the Compactness Theorem. Finally, I will give some interesting applications of compactness, namely the infinite four-colour theorem and the Ax-Grothendieck Theorem.

June 9: Ben Chislett: Ray Tracing and the Light Transport Equation

Abstract:

Ray Tracing is the primary technique for rendering photorealistic images. Advancements in ray tracing techniques have enabled its use in a variety of computer graphics applications, from animated films to real-time video games. In this talk, we explore the Light Transport Equation, Monte Carlo integration, and some of the many optimizations that have led to ray tracing’s rapid rise in popularity.

June 2: Parker Glynn-Adey: The Probabilistic Method

Abstract:

Mathematicians love purity, intuition, and clarity. We like to construct things explicitly with no messiness. But sometimes, mathematical reality is too weird for our explicit methods. Sometimes, we need to look at random examples to find what we’re looking for. In this talk, we use randomness to produce colourings of complete graphs with no monochromatic cliques, a foundational example of the probabilistic method pioneered by Paul Erdős.

Background in counting and graph theory at the MAT A67 / MAT 202 level will help.

May 26: Brian Li: Cops and Robbers with Many Variants

Abstract:

Mathematics and games complement each other in both mathematical research and learning. Cops and Robbers is a game played on graphs between a set of cops and a single robber. The cops begin the game by moving to a set of vertices, with the robber then choosing a vertex to occupy. All players move from vertex-to-vertex along edges. The cops win by successfully occupying the robber’s vertex, hence catching the robber. We will discuss theorems that help us better understand such games and unleash our creativity to explore many different variants. No prior knowledge required.

FAQ

• How long should my talk last?
  • The longest possible talk is 50 minutes, which allows for five minutes for people to arrive and five minutes for questions. Feel free to use up to 50 minutes. Any talk in the range 15 to 50 minutes would be great.
• Do I need to use a particular format for my slides?
  • No, feel free to use any software that you like to make your slides. You don’t even need to make slides. Feel free to draw, or use a whiteboard. For math talks, lots of speakers use Beamer.
• Will my talk be recorded and shared online?
  • By default, yes and yes. The talks are recorded to Zoom Cloud and then Parker edits the recording and puts it on YouTube. If you do not want your talk put online, please let Parker know. He will not post your talk online. We want you to feel comfortable speaking and will accodomate privacy concerns.
• Do I need to present original research?
  • No! Feel free to present anything. We especially appreciate introductory talks, which introduce the audience to an novel topic. If you have original research that you would like to present, then please do so in a friendly and introductory manner.
• Do you have any suggestions for things to present?
  1. Jocelyn R. Bell & Frank Wattenberg (2020) The Slippery Duck Theorem, Mathematics Magazine, 93:2, 91-103, DOI: 10.1080/0025570X.2020.1708693
  2. Sam Chow, Ayla Gafni & Paul Gafni (2021) Connecting the Dots: Maximal Polygons on a Square Grid, Mathematics Magazine, 94:2, 118-124, DOI: 10.1080/0025570X.2021.1869493
  3. Bhargava, Manjul. “The factorial function and generalizations.” The American Mathematical Monthly 107.9 (2000): 783-799.pdf
  4. Birman, Joan S. “New points of view in knot theory.” Bulletin of the American Mathematical Society 28.2 (1993): 253-287. pdf
  5. Margalit, Dan. “The Mathematics of Joan Birman.” Notices of the American Mathematical Society (2019). pdf