“Machine learning you can trust for cosmology and astrophysics”
Prof. Roberto Trotta
(SISSA)

Cosmological and astrophysical data are now sufficiently large and complex to become intractable by traditional statistical techniques. Unravelling the twin mysteries of dark energy and dark matter will require new data analysis methods capable of extracting knowledge from upcoming data streams, including the Euclid satellite, the Nancy Grace Roman space telescope and LSST/Vera Rubin observatory.

In this talk, I will focus on recent advances in simulation-based inference and particularly on neural ratio estimation for amortised marginal posterior inference for fast, scalable inference with guaranteed coverage. I will also present a general, statistically principled solution to the ubiquitous problem of covariate shift in supervised learning, which achieves gold standard performance in a variety of settings, including supernova classification, photometric redshift estimation and weak lensing calibration. I will then give example applications in supernova type Ia cosmology, dark matter direct detection and gravitational waves astronomy.