REU Site: Tiling Theory, Knot Theory, Optimization, Matrix Analysis, and Image Reconstruction

University Of Washington, Seattle WA

Investigators

Abstract

This award is funded in whole or in part under the American Rescue Plan Act of 2021 (Public Law 117-2). This REU Site program will host nine undergraduate students for eight weeks in summers 2022, 2023, and 2024 at the University of Washington Bothell (UWB) for research experiences. Students will provide their research preferences and work closely with a faculty mentor in groups of three. The research projects to be pursued at this REU site are chosen to be of interest to the greater mathematical research community while being accessible to students with minimal background preparation required. The research groups will be strongly encouraged to publish their results in journals and present them at professional meetings. This REU site will prepare its student participants for research careers and graduate training, contribute to increasing diversity in the mathematical sciences, and encourage and prepare its participants to pursue graduate school. The research experience will be complemented by several outings and social activities to build team cohesiveness and networking. This REU Site will focus on research problems in the areas of knot theory, tiling theory, matrix analysis, optimization, and image reconstruction. In the tiling theory group, the participants will experiment with known infinite families of tiles and use basic combinatorial methods to analyze patterns. They will also perform computer-aided explorations of potential minimal aperiodic protosets to determine aperiodicity and potential structure of associated Markov partitions. In the non-smooth optimization group, students will work to improve the numerical performance of the emerging non-smooth spectral gradient methods. This project will also help develop students' scientific computing skills using MATLAB. In the knot theory group, students will explore various open questions, for example, identifying types of knots that arise as components of hexagonal mosaic links created from saturated diagrams and understanding splittable links from specific sizes of saturated hexagonal or parallelogram diagrams. In the medical imaging and optimization group, students will study the algorithmic framework, generate simulated data, and modify existing code to train the deep-learning-based prior and incorporate it into the reconstruction algorithm. This project will offer students the opportunity to gain exposure to cutting-edge concepts in imaging science and machine learning and develop mathematical programming skills. The matrix analysis group will explore open questions on nonnegative matrices, the field of values, the geometry of polynomials, discrete geometry, and number theory. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.