Probabilistic cosmic web classification using fast-generated training data

Abstract

We present a novel method of robust probabilistic cosmic web particle classification in three dimensions using a supervised machine learning algorithm. Training data was generated using a simplified $\Lambda$CDM toy model with pre-determined algorithms for generating halos, filaments, and voids. While this framework is not constrained by physical modeling, it can be generated substantially more quickly than an N-body simulation without loss in classification accuracy. For each particle in this dataset, measurements were taken of the local density field magnitude and directionality. These measurements were used to train a random forest algorithm, which was used to assign class probabilities to each particle in a $\Lambda$CDM, dark matter-only N-body simulation with $256^3$ particles, as well as on another toy model data set. By comparing the trends in the ROC curves and other statistical metrics of the classes assigned to particles in each dataset using different feature sets, we demonstrate that the combination of measurements of the local density field magnitude and directionality enables accurate and consistent classification of halo, filament, and void particles in varied environments We also show that this combination of training features ensures that the construction of our toy model does not affect classification. The use of a fully supervised algorithm allows greater control over the information deemed important for classification, preventing issues arising from hyperparameters and mode collapse in deep learning models. Due to the speed of training data generation, our method is highly scalable, making it particularly suited for classifying large datasets, including observed data.

arXiv comment21 pages, 15 figures. For additional content, see https://github.com/bmbuncher/Prob-CWeb

References in corpus (44)

• Scikit-learn: Machine Learning in Python
• Seven-Year Wilkinson Microwave Anisotropy Probe (WMAP) Observations: Cosmological Interpretation
• The cosmological simulation code GADGET-2
• Simulating the joint evolution of quasars, galaxies and their large-scale distribution
• A Multi-Code Analysis Toolkit for Astrophysical Simulation Data
• The persistent cosmic web and its filamentary structure I: Theory and implementation
• Transients from Initial Conditions in Cosmological Simulations
• Precision Determination of the Mass Function of Dark Matter Halos
• Evolution of the cosmic web
• Tracing the cosmic web
• Transients from Initial Conditions: A Perturbative Analysis
• Bayesian physical reconstruction of initial conditions from large scale structure surveys
• A Dynamical Classification of the Cosmic Web
• Multiscale Phenomenology of the Cosmic Web
• The Multiscale Morphology Filter: Identifying and Extracting Spatial Patterns in the Galaxy Distribution
• NEXUS: Tracing the Cosmic Web Connection
• Universal Density Profile for Cosmic Voids
• A kinematic classification of the cosmic web
• Measurement of Galaxy Cluster Sizes, Radial Profiles, and Luminosity Functions from SDSS Photometric Data
• Precision cosmology with voids: definition, methods, dynamics
• The Cosmic Web, Multi-Stream Flows, and Tessellations
• Galaxy And Mass Assembly (GAMA): The large scale structure of galaxies and comparison to mock universes
• Fast Hamiltonian sampling for large scale structure inference
• ORIGAMI: Delineating Halos using Phase-Space Folds
• The Hierarchical Nature of the Spin Alignment of Dark Matter Haloes in Filaments
• On the connectivity of the cosmic web: theory and implications for cosmology and galaxy formation
• Fast cosmic web simulations with generative adversarial networks
• The tidal evolution of dark matter substructure -- I. Subhalo density profiles
• Machine learning cosmological structure formation
• Automated detection of filaments in the large scale structure of the universe
• Galaxies flowing in the oriented saddle frame of the cosmic web
• T-ReX: a graph-based filament detection method
• Classifying the Large Scale Structure of the Universe with Deep Neural Networks
• The Size of the Longest Filaments in the Universe
• The Persistent Percolation of Single-Stream Voids
• Linearisation with Cosmological Perturbation Theory
• The Size, Shape and Orientation of Cosmological Voids in the Sloan Digital Sky Survey
• Multi-stream portrait of the Cosmic web
• The phase-space structure of nearby dark matter as constrained by the SDSS
• Detection and analysis of cluster-cluster filaments
• Comparing cosmic web classifiers using information theory
• Cosmic web-type classification using decision theory
• Large-scale structures in the $\Lambda$CDM Universe: network analysis and machine learning
• A Machine Learning Approach to Galaxy-LSS Classification I: Imprints on Halo Merger Trees

Cited by in corpus (8)

• Populations of filaments from the distribution of galaxies in numerical simulations
• Hickson-like compact groups inhabiting different environments
• Classification algorithms applied to structure formation simulations
• The four cosmic tidal web elements from the $\beta$-skeleton
• AGNet: Weighing Black Holes with Deep Learning
• Statistical properties of filaments in the cosmic web
• Analysis of Dark Matter Halo Structure Formation in $N$-body Simulations with Machine Learning
• A novel Cosmic Filament catalogue from SDSS data