Seminar Talk Experience and Math and Cookies (Fall 2023)

Math and Cookies is an undergraduate research seminar. If you are interested in giving a talk, please let me know!

The following is the list of tentative speakers for the seminar for Fall 2023. We meet on Mondays 2:00 pm - 3:00 pm.  


Informal meeting with students and Faculty members; David Hobby, Jeungeun Park, Cheyne Glass

This is the first meeting of the semester. Several faculties will join to discuss what mathematics research means to each of us. 

Speaker: Jaiung Jun (SUNY New Paltz)

Title: The space of triangles is a triangle. 

Abstract:  By considering a collection of geometric structures satisfying some conditions, one often obtains another geometric structure, which may provide new perspective on geometric structures of interest. In this talk, I will provide one explicit example, namely a collection of triangles in the plane. 

Speaker: Eric Brattain-Morrin (VideaHealth)

Title: Understanding the New Age of AI: Transformers, Language Models, and Multimodal Learning 

Abstract:  We will explore the exciting world of advanced Artificial Intelligence, introducing powerful models that enable computers to understand and generate human-like text and even interpret images. Transformers, first introduced in 2017's "Attention Is All You Need" for machine translation, have become the backbone of modern AI applications like ChatGPT. We will discuss how Large Language Models (LLMs) leverage transformers to understand and generate human-like text, as well as how these have been extended to process both language and image data. Through this talk, we aim to provide an intuitive understanding of these advanced models, their workings, and their potential applications, so a math background is not necessary. 

Speaker: Chris Eppolito (The University of the South)

Title: Lines and the Pythagorean Theorem 

Abstract:  Using some points in the plane, real number "weights" relating them, and the Pythagorean Theorem, we can draw an interesting collection of lines in the plane. Using high school geometry, we show how to build these lines and ask questions about how they intersect---if time permits, we will see how wiggling the points affects the answers. 

Speaker: Leah Glass (Stack Overflow)

Title: Signals & noise: how to find a (math) job and the skills you need to get there 

Abstract:  When you google “jobs for math majors”, you’ll likely find some common jobs like “statistician”, “actuary” or “math teacher”. And while all of those are great careers, there are dozens of other jobs that need quantitative-minded and -trained employees that have less straightforward (and sometimes, confusingly named!) job titles. What even is a product manager or a platform engineer? This talk will help students understand the variety of jobs and careers that math & engineering majors would be well suited for, as well as highlight other important skills in the job seeking and career navigation process such as interpreting job descriptions, highlighting non-math skills, and tips for finding the right next step after college. 

Speaker: Brian Choi (USMA West Point)

Title: Application of Fractional Calculus to Physics Via Differential Equation 

Abstract:  Physical phenomena are described by Partial Differential Equations (PDE). One of the earliest historical events that marked the beginning of fractional calculus came from Leibniz with the question "what does it mean to take a half-order derivative"? Fractional calculus has been an underdeveloped sub-field of analysis for a long time due to its computational difficulties. With the advent of high-power computing and its applicability to physics, Fractional Calculus has made it come-back. This talk will give an overview of fractional nonlinear Schrodinger equation and its variants by presenting the minimum theoretical background with numerical experiments. 

Speaker: Jeffrey Tolliver (Zoox) 

Title: Option Pricing and Robotic State Estimation 

Abstract:  I will discuss two (unrelated) applications of mathematics in industry: option pricing and robotic state estimation.  Option pricing is based in probability and leads to computational problems that require a variety of classical numerical analysis methods.  State estimation in robotics involves a combination of 3-dimensional geometry and sparse nonlinear least squares problems over manifolds; I will primarily focus on the latter. 

Speaker: Seung-Wook Kim (Bentley University)

Title: Uncovering Consumer Heterogeneity in Big Data: A Hybrid Marketing Science – Deep Learning Approach 

Abstract:  Traditional choice models for understanding market segments are computationally expensive, hampering calibration in Big Data. A potential solution lies in machine learning; however, its lack of marketing theory foundation and interpretability hinders managerial use. Given these challenges, this research describes an approach to the calibration of a choice model with unobserved consumer heterogeneity in a Big Data context. This innovative modeling approach offers promising avenues for large retailers and researchers grappling with conventional marketing science models in the realm of Big Data.