TY - JOUR
T1 - Physics-informed neural networks in the recreation of hydrodynamic simulations from dark matter
AU - Dai, Zhenyu
AU - Moews, Ben
AU - Vilalta, Ricardo
AU - Davé, Romeel
N1 - Funding Information:
ACKNOWLEDGEMENTS This work received support from the Hewlett Packard Enterprise Data Science Institute at the University of Houston. BM acknowledges support from a McWilliams Fellowship at Carnegie Mellon Uni- versity. SIMBA was run on the DiRAC@Durham facility managed by the Institute for Computational Cosmology on behalf of the STFC DiRAC HPC Facility. The equipment was funded by BEIS (Department for Business, Energy & Industrial Strategy) capital funding via STFC capital grants ST/P002293/1, ST/R002371/1, and ST/S002502/1, Durham University, and STFC operations grant ST/R000832/1. DiRAC is part of the National e-Infrastructure.
PY - 2024/1/1
Y1 - 2024/1/1
N2 - Physics-informed neural networks have emerged as a coherent framework for building predictive models that combine statistical patterns with domain knowledge. The underlying notion is to enrich the optimization loss function with known relationships to constrain the space of possible solutions. Hydrodynamic simulations are a core constituent of modern cosmology, while the required computations are both expensive and time-consuming. At the same time, the comparatively fast simulation of dark matter requires fewer resources, which has led to the emergence of machine learning algorithms for baryon inpainting as an active area of research; here, recreating the scatter found in hydrodynamic simulations is an ongoing challenge. This paper presents the first application of physics-informed neural networks to baryon inpainting by combining advances in neural network architectures with physical constraints, injecting theory on baryon conversion efficiency into the model loss function. We also introduce a punitive prediction comparison based on the Kullback-Leibler divergence, which enforces scatter reproduction. By simultaneously extracting the complete set of baryonic properties for the simba suite of cosmological simulations, our results demonstrate improved accuracy of baryonic predictions based on dark matter halo properties and successful recovery of the fundamental metallicity relation, and retrieve scatter that traces the target simulation's distribution.
AB - Physics-informed neural networks have emerged as a coherent framework for building predictive models that combine statistical patterns with domain knowledge. The underlying notion is to enrich the optimization loss function with known relationships to constrain the space of possible solutions. Hydrodynamic simulations are a core constituent of modern cosmology, while the required computations are both expensive and time-consuming. At the same time, the comparatively fast simulation of dark matter requires fewer resources, which has led to the emergence of machine learning algorithms for baryon inpainting as an active area of research; here, recreating the scatter found in hydrodynamic simulations is an ongoing challenge. This paper presents the first application of physics-informed neural networks to baryon inpainting by combining advances in neural network architectures with physical constraints, injecting theory on baryon conversion efficiency into the model loss function. We also introduce a punitive prediction comparison based on the Kullback-Leibler divergence, which enforces scatter reproduction. By simultaneously extracting the complete set of baryonic properties for the simba suite of cosmological simulations, our results demonstrate improved accuracy of baryonic predictions based on dark matter halo properties and successful recovery of the fundamental metallicity relation, and retrieve scatter that traces the target simulation's distribution.
KW - galaxies: evolution
KW - galaxies: haloes
KW - methods: analytical
KW - methods: statistical
UR - http://www.scopus.com/inward/record.url?scp=85179003398&partnerID=8YFLogxK
U2 - 10.1093/mnras/stad3394
DO - 10.1093/mnras/stad3394
M3 - Article
AN - SCOPUS:85179003398
SN - 0035-8711
VL - 527
SP - 3381
EP - 3394
JO - Monthly Notices of the Royal Astronomical Society
JF - Monthly Notices of the Royal Astronomical Society
IS - 2
ER -