François Lanusse
CNRS Researcher @ AIM, CEA Paris-Saclay
Project led by Michael McCabe, Bruno Régaldo, Liam Parker, Ruben Ohana, Miles Cranmer
Accepted at NeurIPS 2024, Best paper award at the NeurIPS 2023 AI4Science Workshop
Navier-Stokes
Incompressible
Compressible
Shallow Water
Diffusion-Reaction
Takamoto et al. 2022
Can we improve performance of surrogate models by pretraining on large quantities of easily simulatable systems?
Normalized MSE:
Context size: 16 frames
M = 0.1
M = 1.0
PDEBench
Presented at NeurIPS 2024 Datasets & Benchmark Track
Flatiron Institute's Raphael Cluster
Pain points
Perfomance optimization
Transposing these methodologies to scientific data and problems brings
unique challenges
Credit: Melchior et al. 2021
Credit:DESI collaboration/DESI Legacy Imaging Surveys/LBNL/DOE & KPNO/CTIO/NOIRLab/NSF/AURA/unWISE
with extensive support from the rest of the team.
Project led by:
Francois
Lanusse
Liam
Parker
Jeff
Shen
Tom
Hehir
Ollie
Liu
Lucas
Meyer
Leopoldo
Sarra
Sebastian Wagner-Carena
Helen
Qu
Micah
Bowles
Presented at NeurIPS 2024 Datasets & Benchmark Track
Ground-based imaging from Legacy Survey
Space-based imaging from JWST
(Blanco Telescope and Dark Energy Camera.
Credit: Reidar Hahn/Fermi National Accelerator Laboratory)
(Subaru Telescope and Hyper Suprime Cam. Credit: NAOJ)
(Dark Energy Spectroscopic Instrument)
(Sloan Digital Sky Survey. Credit: SDSS)
(Gaia Satellite. Credit: ESA/ATG)
Field Embedding Strategy Developed for
Multiple Physics Pretraining (McCabe et al. 2023)
DES g
DES r
DES i
DES z
HSC g
HSC r
HSC i
HSC z
HSC y
(credit)
Survey translation
Spectrum super-resolution
Adaptation at low cost
with simple strategies:
Trained on ->
Eval on ->
Inputs:
measured fluxes
Inputs:
measured fluxes + image
Segmenting central bar and spiral arms in galaxy images based on Galaxy Zoo 3D
Polymathic's recipe for developing Multimodal Scientific Models
Engagement with Scientific Communities
Data Curation And Aggregation
Dedicated ML R&D
Thank you for listening!