Some new developments on sum-of-exponential approximations motivated by physics applications

In this talk, I will discuss some new developments on sum-of-exponential (SOE) approximations, which is motivated by applications in many-body physics. We will examine what could be proved for the complexity of these approximations, emphasizing how it depends on the approximation domain, smoothness in the Fourier space, and the desired accuracy. I will also discuss how physically motivated constraints have led to new algorithmic developments for such a classic problem in numerical analysis.

This seminar is part of the Recent Progress and Open Directions in Matrix Computations series.