Office: ECOT 649
Email: adrianna.gillman (at) colorado.edu
Office hours: By appointment
I have a new webpages. Check it out.
Office: ECOT 649
Email: adrianna.gillman (at) colorado.edu
Office hours: By appointment
I have a new webpages. Check it out.
Roughly speaking, my research interests lie in the intersection of scientific computing, numerical methods for linear PDEs, and numerical linear algebra. The goal of my work (at a high level) is to design efficient and accurate numerical methods of simulating physical phenomena. To achieve this goal, I exploit the physics of the problems under consideration, develop high order discretizations and low rank matrices. My recent work has focused on the following areas:
• “Fast” algorithms such as the Fast Multipole Method and fast direct solvers.
• Integral equation formulations for the mathematical modeling of physics and other real world applications.
• Wave scattering, acoustics, variable media.
• Techniques for handling challenging geometries such as corners, edges, and periodic domains.
The methods I develop can be applied to a variety of problems including the simulation of composite materials and design of microfluidic devices.
librla: A randomized linear algebra library has been released. It runs natively in Python, Matlab and Julia. It is efficient and has more versatilitiy than other randomized linear algebra packages. A manuscript has been submitted for publication.
A. Gillman is participating in the semester long program on Interfaces and Unfitted Discretization Methods at the Mittag-Leffler Institute.
A. Gillman will spend her sabbatical year as a Guest Researcher at KTH with the support of a Knut and Alice Wallenberg Foundation grant.
1 and 2D HPS demo codes are available on GitHub