The projects running in Fall 2024 will be listed below.

**Important Dates:**

Friday, August 9, 2024 – Faculty Project Proposals Due

Friday, August 16, 2024 – Undergraduate and Graduate Applications Open

* *You must use your wisc.edu account to access the application forms **

Monday, August 26, 2024 – Undergraduate and Graduate Applications Due

Wednesday, September 4, 2024 – All teams will have been formed and all applicants informed of their status

This is an accordion element with a series of buttons that open and close related content panels.

## Computing / visualizing Patterson-Sullivan measures on the circle

Given a discrete group of symmetries acting on a negatively-curved space X, there is a class of measures (“Patterson–Sullivan measures”) on the boundary of X which reflects the data of that group action. The objective of this somewhat exploratory project will be to compute and visualize some Patterson–Sullivan measures.

We will concentrate on actions of free groups, surface groups, or Coxeter groups on the hyperbolic plane or on convex domains in the real projective plane. In this case, the boundary of X, where the Patterson–Sullivan measures live, is homeomorphic to a circle, and there are existing Python packages that can be used.

**Faculty Mentor:** Feng Zhu

**Graduate Student:** Ron Yang

**Project Level:** Intermediate

**Recommended Undergraduate Courses or Experience:**

- Linear Algebra
- Python

## Distribution of random vectors coming from geometry

This project is about distribution of vectors coming from translation surfaces which are shapes that locally look like R^2, outside of finitely many cone points. In previous semesters, we have done lots of experimental work to generate conjectures, and now we are in a position to prove some theorems and generate more conjectures! We are looking for 2-3 undergraduate students who are proficient in python to improve the current code, and finish the project. The advanced mathematical background can be “blackboxed” so that one can only focus on computing, linear algebra, and multivariable calculus.

**Faculty Mentor: **Caglar Uyanik

**Project Level: **Intermediate

**Recommended Undergraduate Courses or Experience:**

- Linear algebra and multivariable calculus
- Upper intermediate programming skills

## Entropy calculation for non Archimedean dynamics

My collaborators and I recently figured out an algorithm for the calculation of the inverse of the Artin-Mazur zeta function for 1d rational maps on locally compact non Archimedean fields. We are interested in implementing it as well as optimizing it to cut down the time complexity.

**Faculty Mentor: **Chenxi Wu

**Graduate Student: **TBD – Accepting Applications

**Project Level:** Advanced – students who have taken multiple upper-level mathematics courses

**Recommended Undergraduate Courses or Experience:**

- Familiarity with linear algebra, rings and fields, and metric spaces
- Some basic knowledge of programming would be good!

## Hyperbolic geometry through crochet

This project is exploratory in nature, its main goal is to understand and demonstrate the unique properties of hyperbolic geometry through the medium of crochet. Hyperbolic geometry, unlike Euclidean geometry, is defined by its constant negative curvature, leading to interesting shapes akin to lettuce leaves. As we crochet these models other avenues of investigation will be explored depending on the interest of the team members.

**Faculty Mentor: **Grace Work

**Graduate Student: **TBD – Accepting Applications

**Project Level:** Beginner

**Recommended Undergraduate Courses or Experience:**

- Students should have completed 234
- Experience with crochet techniques

## Kayles game on graphs

The project explores a periodicity of Kayles game on graphs.

We will start with computational aspects of classical Kayles game. Kayles is a simple impartial game played with rows of pins. Each player, on his or her turn, may remove either any one pin, or two adjacent pins. This game is periodic, which can be shown only after computations, which we will provide.

On the second part we will extend our tools to the game on some graphs’ families. If we think of the classical rows of pins as path graphs, then we can have game on stars, caterpillars, general trees, etc.

**Faculty Mentor: **Mikhail Ivanov

**Graduate Student: **TBD – Accepting Applications

**Project Level:** Beginner

**Recommended Undergraduate Courses or** **Experience:**

- Minimal knowledge of graph theory’s language is recommended (CS/Math 240 is enough)
- Minimal python skills are required

## Network regression via supervised latent motifs

The goal of this project is to develop a framework for network-level regression, two make sense of approximating an unknown network X by a “linear combination” of a set of known networks for assessing structural similarity and application in graph classification problems. During summer REU 2024, we have developed a method for network regression by combining k-path subgraph sampling and supervised matrix factorization. We would like to continue developing this method and benchmark its performance on graph classification problems with popular graph neural network methods.

**Faculty Mentor:** Hanbaek Lyu

## Numerical study of Kakeya maximal inequalities

The aim of the project is to find some numerical evidence for the Kakeya maximal function conjecture in 3 dimensions. This conjecture roughly states that a collection of tubes in 3-space of equal radii and separated directions is mostly disjoint. It is folklore that if the placement of each tube is random (uniform), then the conjecture is true, so we will first aim to verify this numerically (and possibly theoretically). Then, we will try to find a better way to search parameter space for counterexamples to test whether 3/2 is the correct exponent. We may also try to develop some of these approaches in 2 dimensions first, where the conjecture is known.

**Faculty Mentor:** Terence Harris

**Graduate Student: **Kaiyi Huang

**Project Level: **Intermediate

**Recommended Undergraduate Courses or Experience:**

- Math 234 (or some knowledge of multivariable calculus).

## Solving Partial Differential Equations Through Neural Network Based Methods

This project explores innovative approaches to solving partial differential equations (PDEs) using neural network-based methods. PDEs are fundamental in describing various phenomena in physics, engineering, and other fields, but traditional analytical and numerical solutions can be complex and computationally intensive. Neural networks, with their ability to approximate complex functions, offer a promising alternative for solving PDEs more efficiently.

The primary goal of this project is to develop and implement neural network models to solve different types of PDEs, such as heat equations, wave equations, and fluid dynamics equations. Students will learn how to design, train, and evaluate these models, comparing their performance with traditional methods. By the end of the program, participants will gain hands-on experience in applying machine learning techniques to classical mathematical problems, preparing them for advanced research and practical applications in computational science.

**Faculty Mentor: **Yukun Yue

**Graduate Student: **TBD – Accepting Applications

**Project Level:** Advanced – students who have taken multiple upper-level mathematics courses

**Recommended Undergraduate Courses or Experience:**

- MATH 321 or MATH 322 or some other equivalent courses
- Programming experience in python