florpi
https://florpi.github.io/
IAIFI Fellow
Carol(ina) Cuesta-Lazaro
1. Observations x Simulations
2. Learning a general parametrisation for feedback
Constrained Simulations
(IC reconstruction)
3. Compressing simulation snapshots in continuous time
Inverse modelling
Data driven subgrid models
1 to Many:
Distribution of Galaxies
Underlying Dark Matter
["Debiasing with Diffusion: Probabilistic reconstruction of Dark Matter fields from galaxies"
Ono et al (including Cuesta-Lazaro) arXiv:2403.10648]
Victoria Ono
Core Park
TNG-300
True DM
Inferred DM
Size of training simulation
Galaxy Cluster
Void
[arXiv:2403.10648]
Model trained on Astrid subgrid model
["3D Reconstruction of Dark Matter Fields with Diffusion Models: Towards Application to Galaxy Surveys" Park, Mudur, Cuesta-Lazaro et al (in-prep)]
Posterior Sample
Posterior Mean
Guided simulations with fuzzy constraints
1. Predictive Simulators learned subgrid models from high resolution simulations
Hydro simulator
Subgrid model
Solution: train on the fly
2. Data-driven subgrid models learned subgrid models from high dimensional observations
Hydro simulator
Subgrid model
What is the space of plausible solutions and how do we search it?
Are these models predictive?
[Image credit: Sarah Jeffreson's beautiful high res sims]
["Multifield Cosmology with Artificial Intelligence" Villaescusa-Navarro et al arXiv:2109.09747]
Out-of-Distribution
In-Distribution
Informative abstractions of the data
Transfer learning beyond LCDM
Cosmic web Anomaly Detection
Representing baryonic feedback
Contrastive
Generative
inductive biases
from scratch or from partial observations
Students at MIT are
OVER-CAFFEINATED
NERDS
SMART
ATHLETIC
Simulator 1
Simulator 2
Dark Matter
Feedback
i) Contrastive
Baryonic fields
ii) Generative
Baryonic fields
Dark Matter
Generative model
Total matter, gas temperature,
gas metalicity
Encoder
~ 10 trillion particles per snapshot stored
x Discrete snapshots