Predicting Localized Primordial Star Formation with Deep Convolutional Neural Networks

Abstract

We investigate applying 3D deep convolutional neural networks as fast surrogate models of the formation and feedback effects of primordial stars in hydrodynamic cosmological simulations of the first galaxies. Here, we present the surrogate model to predict localized primordial star formation; the feedback model will be presented in a subsequent paper. The star formation prediction model consists of two sub-models: the first is a 3D volume classifier that predicts which (10 comoving kpc)${}^3$ volumes will host star formation, followed by a 3D Inception-based U-net voxel segmentation model that predicts which voxels will form primordial stars. We find that the combined model predicts primordial star forming volumes with high skill, with $F_1>0.995$ and true skill score $>0.994$. The star formation is localized within the volume to $\lesssim 5^3$~voxels ($\sim 1.6$~comoving kpc${}^3$) with $F_1>0.399$ and true skill score $>0.857$. Applied to simulations with low spatial resolution, the model predicts star forming regions in the same locations and at similar redshifts as sites in resolved full-physics simulations that explicitly model primordial star formation and feedback. When applied to simulations with lower mass resolution, we find that the model predicts star forming regions at later redshift due to delayed structure formation resulting from lower mass resolution. Our model predicts primordial star formation without halo finding, so will be useful in spatially under-resolved simulations that cannot resolve primordial star forming halos. To our knowledge, this is the first model that can predict primordial star forming regions that match highly-resolved cosmological simulations.

arXiv comment18 pages, 13 figures

References in corpus (35)

• Adam: A Method for Stochastic Optimization
• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Planck 2013 results. XVI. Cosmological parameters
• Deep Residual Learning for Image Recognition
• Densely Connected Convolutional Networks
• Going Deeper with Convolutions
• Properties of galaxies reproduced by a hydrodynamic simulation
• A Multi-Code Analysis Toolkit for Astrophysical Simulation Data
• The Circumgalactic Medium
• Enzo: An Adaptive Mesh Refinement Code for Astrophysics
• Multi-scale initial conditions for cosmological simulations
• The Birth of a Galaxy: Primordial Metal Enrichment and Stellar Populations
• HOP: A New Group-Finding Algorithm for N-body Simulations
• Formation of the First Stars
• How To Model Supernovae in Simulations of Star and Galaxy Formation
• Galaxy Properties and UV Escape Fractions During Epoch of Reionization: Results from the Renaissance Simulations
• The Destruction of Cosmological Minihalos by Primordial Supernovae
• Be it therefore resolved: Cosmological Simulations of Dwarf Galaxies with Extreme Resolution
• Enzo+Moray: Radiation Hydrodynamics Adaptive Mesh Refinement Simulations with Adaptive Ray Tracing
• The Birth of a Galaxy. II. The Role of Radiation Pressure
• The First Population II Stars Formed in Externally Enriched Mini-halos
• Classifying Radio Galaxies with Convolutional Neural Network
• Connecting the first galaxies with ultra faint dwarfs in the Local Group: chemical signatures of Population~III stars
• Metal Cooling in Simulations of Cosmic Structure Formation
• Radiative Stellar Feedback in Galaxy Formation: Methods and Physics
• U-Net: Convolutional Networks for Biomedical Image Segmentation
• How the First Stars Regulated Local Star Formation I: Radiative Feedback
• Compressed Convolutional LSTM: An Efficient Deep Learning framework to Model High Fidelity 3D Turbulence
• The Deaths of Very Massive Stars
• From Dark Matter to Galaxies with Convolutional Networks
• Building high accuracy emulators for scientific simulations with deep neural architecture search
• Efficient Object Localization Using Convolutional Networks
• CosmoFlow: Using Deep Learning to Learn the Universe at Scale
• External Enrichment of Minihalos by the First Supernovae
• Combo Loss: Handling Input and Output Imbalance in Multi-Organ Segmentation

Cited by in corpus (3)

• Supervised machine learning on Galactic filaments Revealing the filamentary structure of the Galactic interstellar medium
• The problem of dust attenuation in photometric decomposition of edge-on galaxies and possible solutions
• Connecting Primordial Star Forming Regions and Second Generation Star Formation in the Phoenix Simulations