- Department of Data Science
I am a mathematician working in numerical analysis, with a focus on dynamical low-rank approximation and its applications in scientific computing and machine learning. In my research, I develop low-rank neural networks, where weight matrices are represented in low-rank form. A key aspect of this work is the design of geometry-aware training algorithms that respect the differential geometry of the low-rank matrix manifold. This approach enables the development of robust training methods with provable convergence to low-rank optima.
In the realm of scientific computing, my work centers on efficient numerical methods for high-dimensional problems in computational quantum mechanics and radiation transport, where dynamical low-rank methods play a crucial role in reducing computational complexity.
Areas of Work
Publications