Fast inference of Cosmology from High Resolution Maps using Deep Learning - Arne Thomsen, ETH Zurich
Ongoing galaxy surveys observe the large-scale structure of the Universe. Conventionally, constraints on the cosmological parameters are calculated by comparing two-point functions of the observables with semi-analytical theory predictions. However, we know that due to nonlinear structure formation at late times, the physical fields contain information which cannot be captured in this way. In this talk, I present an HPC pipeline to leverage numerical simulations and the expressive power of deep learning to extract this additional cosmological information by learning the summary statistic instead.