Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 314 - Statistical Challenges in Cosmology
Type: Topic-Contributed
Date/Time: Wednesday, August 11, 2021 : 3:30 PM to 5:20 PM
Sponsor: Section on Nonparametric Statistics
Abstract #317654
Title: Learning to Predict Structure Formation in the Universe
Author(s): Yin Li*
Companies: Flatiron Institute
Keywords: Cosmology; Deep learning; Simulation; Surrogate model; Super resolution; Style transfer

On the largest scales, our Universe features filamentary structures known as the cosmic web. They formed under the highly nonlinear gravitational collapse, from simple initial conditions. Physics encoded in the initial conditions and the growth history can be inferred from data observed by ground- and space-based telescopes. Structure formation models are needed for such analyses, for which expensive numerical simulations are usually performed. We developed neural network surrogate models that are both fast and accurate. We introduced global dependence in the network on cosmological parameters, as continuous styles. Furthermore, we built GAN-based networks that can enhance the mass resolutions of simulations by a factor of 512. Our models generalize surprisingly well in many cases. With further development, they have the potential to maximize the scientific return from the current and future cosmological observations.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2021 program