Probabilistic cosmic web classification using fast-generated training data. Issue 4 (13th July 2020)

• Record Type: Journal Article
• Authors: Buncher, Brandon; Carrasco Kind, Matias
• Is Part Of: Monthly notices of the Royal Astronomical Society. Volume 497:Issue 4(2020)
• Page Start: 5041
• Page End: 5060
• Publication Date: 2020-07-13
• Subjects: methods: data analysis -- methods: statistical -- galaxies: evolution -- galaxies: haloes -- large-scale structure of Universe -- dark matter; Astronomy -- Periodicals; Periodicals; 520.5
• DOI: 10.1093/mnras/staa2008
• Languages: English
• ISSNs: 0035-8711

Abstract

ABSTRACT: We present a novel method of robust probabilistic cosmic web particle classification in three dimensions using a supervised machine learning algorithm. Training data were generated using a simplified ΛCDM toy model with pre-determined algorithms for generating haloes, filaments, and voids. While this framework is not constrained by physical modelling, it can be generated substantially more quickly than an N-body simulation without loss in classification accuracy. For each particle in this data set, 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 ΛCDM, 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 data set 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 arbitrary 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 data sets, including observed data.