["DESI 2024 VI: Cosmological Constraints from the Measurements of Baryon Acoustic Oscillations" arXiv:2404.03002]
What role did Machine Learning play?
Dark Energy is constant over time
Dataset Size = 1
Can't poke it in the lab
Simulations
Bayesian statistics
1-Dimensional
Machine Learning
Secondary anisotropies
Galaxy formation
Intrinsic alignments
DESI, DESI-II, Spec-S5
Euclid / LSST
Simons Observatory
CMB-S4
Ligo
Einstein
Unicorn land The promise of ML for Cosmology
Reality Check Roadblocks & Bottlenecks
Mapping dark matter
Reverting gravitational evolution
Field Level Inference
Learning to represent baryonic feedback
Data-driven hybrid simulators
Unsupervised problems
[Image Credit: Claire Lamman (CfA/Harvard) / DESI Collaboration]
["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
Long range correlations
Huge pointclouds (20M)
Homogeneity and isotropy
Siddharth Mishra-Sharma
Lesson #1: leverage data representations + symmetries
Diffusion model
CNN
Diffusion
Increasing Noise
["Diffusion-HMC: Parameter Inference with Diffusion Model driven Hamiltonian Monte Carlo"
Mudur, Cuesta-Lazaro and Finkbeiner]
Nayantara Mudur
["Your diffusion model is secretly a certifiably robust classifier"
Chen et al
arXiv:2402.02316]
CNN
Diffusion
Lesson #2: learning likelihoods can be more robust than poseriors
Just use binary classifiers!
Binary cross-entropy
Sample from simulator
Mix-up
Likelihood-to-evidence ratio
["Likelihood-free MCMC with Amortized Approximate Ratio Estimator" Hermans et al]
Lesson #3: Classifiers are awesome
Likelihood-to-evidence ratio
["Do Deep Generative Models know what they don't know?" Nalisnick et al]
p(x)
Classsifier
Simulations
Observation
Lesson #4: What should x be?
Observed
Simulated
1 to Many:
Distribution of Galaxies
Underlying Dark Matter
["Debiasing with Diffusion: Probabilistic reconstruction of Dark Matter fields from galaxies"
Ono et al arXiv:2403.10648]
Victoria Ono
Core Park
Lesson #5: Most problems 1 to Many
Truth
Sampled
Observed
Small
Large
Scale (k)
Power Spectrum
Small
Large
Scale (k)
Cross correlation
TNG-300
True DM
Sample DM
["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
Stochastic Interpolants
NF
Lesson #6: Match two distributions that are already close!
?
["Probabilistic Forecasting with Stochastic Interpolants and Foellmer Processes" Chen et al arXiv:2403.10648 (Figure adapted from arXiv:2407.21097)]
Guided simulations with fuzzy constraints
Can we run larger simulations? (DESI volumes)
At high resolution?
Faster?
All this works depends on simulations, but...
Thousands of them?
Gravitational evolution ODE
Particle-mesh
Particle-mesh
N-body
Hybrid Simulator - on the fly
Gravitational evolution ODE
Trained to match particle velocities and positions: DIFFERENTIABLE
Particle-mesh
N-body
Hybrid ML-Simulator
"Nbodyify: Adaptive mesh corrections for PM simulations" Cuesta-Lazaro, Modi in preps
Lesson #7: Substantial speed ups without accuracy loss are very hard to achieve
What is the space of plausible solutions and how do we search it?
Differentiable Galaxies ODEs
Our best bet
Neural Network corrections
Data-driven hybrid simulators
Are these models predictive?
~ 10 trillion particles per snapshot stored
x Discrete snapshots
Can we learn compressed continuous representations with Neural Fields?
How do we learn what is the robust information?
Simulating dark matter is easy!
"Atoms" are hard" :(
N-body Simulations
Hydrodynamics
Can we improve our simulators in a data-driven way?
(if cold!)
~ Gpc
pc
kpc
Mpc
Gpc
[Video credit: Francisco Villaescusa-Navarro]
Gas density
Gas temperature
Small
Large
In-Distribution
In-Distribution
In-Distribution
Out-of-Distribution
Out-of-Distribution
Out-of-Distribution
Out-of-Distribution
Out-of-Distribution
Out-of-Distribution
["Multifield Cosmology with Artificial Intelligence" Villaescusa-Navarro et al arXiv:2109.09747]
Out-of-Distribution
In-Distribution
Simulator 1
Simulator 2
Dark Matter
Feedback
Contrastive
Lesson #8: Think carefully about the representations you care about
Parity violation cannot be originated by gravity
["Measurements of parity-odd modes in the large-scale 4-point function of SDSS..." Hou, Slepian, Chan arXiv:2206.03625]
["Could sample variance be responsible for the parity-violating signal seen in the BOSS galaxy survey?" Philcox, Ereza arXiv:2401.09523]
Real or Fake?
x or Mirror x?
Train
Test
Me: I can't wait to work with observations
Me working with observations:
Lesson #9: Low data regime + low signal to noise ratio = difficult to find data-efficient architectures
1. There is a lot of information in galaxy surveys that ML methods can access
2. We can tackle high dimensional inference problems so far unatainable
3. Our ability to simulate limits the amount of information we can robustly extract
Hybrid simulators, forward models, robustness
Unsupervised problems: parity violation
Mapping dark matter, constrained simulations... Let's get creative!
Field level inference