Speaker: Erika Ye, MIT PSFC

Abstract: Simulating and understanding the behavior of high-temperature plasmas is an important step towards achieving fusion energy. However, while it is known that turbulent behavior does significantly influence macroscopic dynamics, kinetic simulations of collisionless (or weakly collisional) plasmas are often extremely expensive or infeasible due to high resolution requirements and the exponential scaling of computational cost with respect to dimension. As such, the possibility that quantum computers can provide large (maybe exponential) speed-up is very attractive. In this talk, I will introduce quantum algorithms in the context of plasma physics and present some of the work that has been done at the PSFC. I will also discuss algorithms that are inspired by quantum algorithms but run on classical hardware. These quantum-inspired algorithms formally obtain a similar speed-up, but at the cost of some accuracy. We will conclude by discussing the outlook of alternative computing for further plasma physics.

Bio: Erika was a postdoc at the PSFC but will be starting at Lawrence Berkeley National Lab as an Alvarez fellow. Her research is focused on developing quantum-inspired tensor network algorithms for solving partial differential equations, which rely on making low-rank approximations to perform calculations with reduced computational resources. Though the algorithms themselves are problem agnostic, over the past two years, she has been looking at their performance specifically for solving the collisionless Vlasov-Maxwell’s equations. Erika also dabbles with quantum computing algorithms for solving nonlinear PDEs because of the close connections between the two fields.