Title: Bridging the Weights and Maps: interpreting structure formation in the era of precision cosmology
Abstract: Our concordance cosmology model paints a picture of a Universe in which the total matter content is dominated by an enigmatic dark matter that drives structure formation. Over recent decades, cosmological probes, such as galaxy clustering, weak gravitational lensing, and the cosmic microwave background, have confirmed this picture with remarkable agreement on large scales. As we now enter cosmology’s most ambitious era, we are poised to map the Universe’s structure in unprecedented detail. But precision measurements demand precision interpretation, particularly on smaller scales where astrophysical complexity dominates. Two critical challenges that currently limit our ability to extract unbiased cosmology from large-scale structure surveys are uncertainties in cluster mass estimation and complexities in connecting the observed distribution of galaxies to the underlying dark matter field. These challenges are not independent and must be modeled jointly. High-fidelity cosmological simulations provide powerful laboratories for understanding structure formation and how baryons connect to dark matter, while machine learning captures the non-linear complexity of astrophysical processes shaping observations. In this talk, I highlight successes in using simulations and machine learning to mitigate systematics affecting cluster mass estimation and galaxy bias. I’ll present my vision for building a research program at this intersection that trains the next generation of computational cosmologists and ensures we can deliver on our promise to reveal how structure grows across cosmic time, how dark matter behaves, and what this may imply for our Universe’s future.