Very few assumptions about the mechanisms of galaxy formation
Symmetries: Homogeneous and isotropic Universe on large scales
(On large scales)
Due to being fast
Small Scales
Galaxy properties
Astro parametrization
DM density + velocity fields
Target
MTNG
Astrid
UniverseMachine
CAMELS
Data
Hydro Sims
SAMS
Santa Cruz
Dark Matter
"Baryons"
Transformer
How many parameters can all these simulations share?
MTNG
Astrid
UniverseMachine
CAMELS
Data
Hydro Sims
SAMS
Santa Cruz
Can we fit the remaining to observations directly?
Learning to represent feedback
SIMBA
TNG
Astrid
Different Supernovae and AGN feedback mechanisms
+ Different parametrisations
Finding anomalies
Z
"A point cloud approach to generative modeling for galaxy surveys at the field level"
Cuesta-Lazaro and Mishra-Sharma
arXiv:2311.17141
Base Distribution
Target Distribution
Base Distribution
Denoiser
Invariant
Equivariant
All learnable functions
All learnable functions constrained by your data
All Equivariant functions
More data efficient!
Equivariant models
Non-equivariant
SEGNN: arXiv:2110.02905
NequIP: arXiv:2101.03164
Very few assumptions about the mechanisms of galaxy formation
Symmetries: Homogeneous and isotropic Universe on large scales
(On large scales)
Due to being fast
Small Scales
Learned
Learned representations
Symmetries: Homogeneous and isotropic Universe on large scales
as long as we can sufficiently simulate
Grokking
Double Descent
Emergent abilities
IAIFI Fellow
Carolina Cuesta-Lazaro
Art: "Melancholy" by Edvard Munch
IAIFI Fellow
Carolina Cuesta-Lazaro
Art: "Midlife crisis" by Nik Ad