Computational Statistics and Inference for Black Hole Astronomy

Black Holes are one of the most mysterious objects in astronomy. In the past decade, we have seen an explosion of data-driven black hole research. One such example is the first-ever image of a black hole's horizon from the Event Horizon Telescope in 2019. In this talk, I'll present how astronomers infer a black hole's properties, highlighting the unique, difficult, high-dimensional, and complex geometry inference and sampling problem we face. I'll also discuss some of the unique challenges we face as astronomers, including handling complicated non-identifiable posteriors and inferring parameters from complex systems described by differential equations. Finally, I'll discuss the next frontier of black hole research and how computational statistics and machine learning are critical to future discoveries of these fundamental objects.