Nevada State University Mathematics Colloquium is held monthly on the second or third Tuesday of each month. Typically, colloquium talks are given immediately following our Math Tea. For inquiries, please contact Sungju Moon or Nandita Sahajpal.
Fall 2023
Oct 10: Emily A. Robinson (Cal Poly San Obispo)
Human Perception of Statistical Charts: An Introduction to Graphical Testing Methods
Abstract. In a world full of data, we all consume graphics on a regular basis in order to inform our decisions and aid in making discoveries about the information presented. With the continuous relevance of graphics, it is important for statisticians in any specialty area to understand what makes a chart good or bad. Through testing, we can better inform and establish fundamental principles for creating and displaying graphics. In this presentation, I will first motivate the talk with a brief history of graphics and established data visualization studies. I will then introduce recent methods used for testing graphics and share my work on establishing guidelines for the use of logarithmic scales.
Nov 14: Taylor McAdam (Pomona College)
Chaos on the Circle
Abstract. Rotate a circle by a fixed angle, then repeat again and again. Where will a single point travel? Will it come back to where it started and how does the answer depend on the rotation angle? Despite their simplicity, rotations and other transformations of the circle teach us a lot about many processes like planets orbiting a sun, coffee stirred in a cup, and about the very nature of numbers themselves. In this talk, I will discuss a variety of ways that mathematicians think about how “complicated" or chaotic such a process becomes. Along the way we find surprising connections to other areas of math such as (ir)rationality of real numbers and binary sequences.
Nov 21: Xinyu Zhao (McMaster U.)
Instability of the 2D Euler Equations
Abstract. Hydrodynamic stability is a well-established but still highly active research area in fluid dynamics, with pioneering work by Helmholtz, Kelvin, Rayleigh and Reynolds during the nineteenth century. To develop an understanding of how “small” perturbations grow and modify fluid flows, one usually linearizes the governing equations, and investigates the eigenvalues and eigenvectors of the linearized operator. The flow is considered linearly unstable if the operator possesses an eigenvalue with a positive real part. In this talk, we will present a numerical study of the instability of the 2D Taylor-Green vortex, which is a steady solution of the 2D Euler equations. This is based on a joint work with Bartosz Protas and Roman Shvydkoy.
Spring 2024
Feb 13: Andrew Lavengood-Ryan (NV State)
Emergence of Linear Diophantine Equations in DNA Self-Assembly
Abstract. In this talk, we’ll explore the fundamentals of DNA self-assembly and see how graph theory can be used to model this biological process. As we explore the structure of the complexes that can be produced from this process, we will see how a number-theoretic structure - in particular, linear diophantine equations - emerge naturally. We will also see how this structure makes predicting the size of the DNA complexes a simple gcd computation.
Mar 12: Julie Vega (Maret School)
Braid Group Cryptography
Abstract. Think briefly about braiding hair or a friendship bracelet. While forming your beautiful creation you are changing the position of each string (or piece of hair) in the process. Intuitively, the braid group is a set of elements that comes from recording the position of strings throughout the braiding process. The braid group is a fascinating group and can be found in many contexts including quantum physics, robotics, and on the surface of the sun! This talk will introduce the braid group and investigate some group properties. We will end the talk with a look at how the braid group can be used in cryptography, a field interested in sending and receiving private encrypted messages.
Apr 16: Rachel Petrik (Rose-Hulman)
Close Encounters of the $K$-KInd: Navigating the Neighborhoods of KNN Classifiers and Regressors
Abstract. In this talk, we’ll dive into the world of $K$-Nearest Neighbors (KNN), a straightforward yet powerful machine-learning algorithm that relies on the concept of proximity for both classification and regression tasks. ``Close Encounters of the $K$-Kind’’ takes you on a journey to decode the essentials of KNN and understand how `neighborly’ data points contribute to predictions. In addition, we’ll offer strategies to overcome common pitfalls like the curse of dimensionality and differing feature units.
Click here to view recordings of the past colloqium talks.