Choose timezone
Your profile timezone:
Simulations-based inference (SBI) has emerged as a powerful tool for extracting significantly more cosmological information from large-scale structure probes. However, building SBI is a complex and challenging task, requiring interdisciplinary work in map-level modeling (substructure, galaxy-halo connection, galaxy evolution, baryon feedback, intrinsic alignments), high-performance computing (building N-body simulators, parallel code optimization, dealing with IO and storage), and artificial intelligence (map-level emulation, painting, cosmological inference methodology). In this talk, I will discuss key elements of these aspects and present how we can build SBI for Stage-IV surveys in practice. I will also share our insights from the previous and ongoing map-level analyses of the Dark Energy Survey.
Martine Lokken, Jonás Chaves Montero