Initial Conditions of the Universe

ML for Large Scale Structure:

A wish list

Forward diffusion: Add Gaussian noise (fixed)

Diffusion models

Diffusion on 6D coordinates

Reverse diffusion: Denoise previous step

Forward diffusion: Add Gaussian noise (fixed)

Setting tight constraints with only 5000 galaxies

All learnable functions

Equivariant functions

Data constraints

Credit: Adapted from Tess Smidt

Equivariant diffusion

Implications for robustness and interpretability?

+ Galaxy formation

+ Observational systematics (Cut-sky, Fiber collisions)

Forward Model

N-body simulations

Observations