Approximate Bayesian Uncertainties on Deep Learning Dynamical Mass Estimates of Galaxy Clusters

Abstract

We study methods for reconstructing Bayesian uncertainties on dynamical mass estimates of galaxy clusters using convolutional neural networks (CNNs). We discuss the statistical background of approximate Bayesian neural networks and demonstrate how variational inference techniques can be used to perform computationally tractable posterior estimation for a variety of deep neural architectures. We explore how various model designs and statistical assumptions impact prediction accuracy and uncertainty reconstruction in the context of cluster mass estimation. We measure the quality of our model posterior recovery using a mock cluster observation catalog derived from the MultiDark simulation and UniverseMachine catalog. We show that approximate Bayesian CNNs produce highly accurate dynamical cluster mass posteriors. These model posteriors are log-normal in cluster mass and recover $68\%$ and $90\%$ confidence intervals to within $1\%$ of their measured value. We note how this rigorous modeling of dynamical mass posteriors is necessary for using cluster abundance measurements to constrain cosmological parameters.

arXiv comment15 pages, 5 figures. 3 tables, accepted for publication at ApJ

References in corpus (30)

• Adam: A Method for Stochastic Optimization
• Planck 2013 results. XVI. Cosmological parameters
• Neural Architecture Search with Reinforcement Learning
• Machine learning and the physical sciences
• The Rockstar Phase-Space Temporal Halo Finder and the Velocity Offsets of Cluster Cores
• Weight Uncertainty in Neural Networks
• UniverseMachine: The Correlation between Galaxy Growth and Dark Matter Halo Assembly from z=0-10
• Cosmological Parameters from Observations of Galaxy Clusters
• MultiDark simulations: the story of dark matter halo concentrations and density profiles
• Tracing cosmic evolution with clusters of galaxies
• Bayesian Convolutional Neural Networks with Bernoulli Approximate Variational Inference
• Planck 2015 results. XXIV. Cosmology from Sunyaev-Zeldovich cluster counts
• Weighing the Giants IV: Cosmology and Neutrino Mass
• MAMPOSSt: Modelling Anisotropy and Mass Profiles of Observed Spherical Systems. I. Gaussian 3D velocities
• CMU DeepLens: Deep Learning For Automatic Image-based Galaxy-Galaxy Strong Lens Finding
• SuperNNova: an open-source framework for Bayesian, Neural Network based supernova classification
• A Deep Learning Approach to Galaxy Cluster X-ray Masses
• Deeply Uncertain: Comparing Methods of Uncertainty Quantification in Deep Learning Algorithms
• Interloper treatment in dynamical modelling of galaxy clusters
• A Machine Learning Approach for Dynamical Mass Measurements of Galaxy Clusters
• Galaxy Cluster Mass Estimation from Stacked Spectroscopic Analysis
• A Robust and Efficient Deep Learning Method for Dynamical Mass Measurements of Galaxy Clusters
• Dynamical Mass Measurements of Contaminated Galaxy Clusters Using Machine Learning
• Prediction of galaxy halo masses in SDSS DR7 via a machine learning approach
• Galaxy Cluster Mass Reconstruction Project: III. The impact of dynamical substructure on cluster mass estimates
• Dynamical mass inference of galaxy clusters with neural flows
• Galaxy Cluster Mass Reconstruction Project - IV. Understanding the effects of imperfect membership on cluster mass estimation
• GalWeight: A New and Effective Weighting Technique for Determining Galaxy Cluster and Group Membership
• The XXL Survey: XXIII. The Mass Scale of XXL Clusters from Ensemble Spectroscopy
• Cosmic Visions Dark Energy: Science

Cited by in corpus (22)

• The DAWES review 10: The impact of deep learning for the analysis of galaxy surveys
• Simulation-Based Inference of Reionization Parameters From 3D Tomographic 21 cm Lightcone Images
• A Deep Learning Approach to Infer Galaxy Cluster Masses from Planck Compton$-y$ parameter maps
• The scatter in the galaxy-halo connection: a machine learning analysis
• Simulation-based inference of dynamical galaxy cluster masses with 3D convolutional neural networks
• The Dynamical Mass of the Coma Cluster from Deep Learning
• What to expect from dynamical modelling of cluster haloes II. Investigating dynamical state indicators with Random Forest
• Machine Learning methods to estimate observational properties of galaxy clusters in large volume cosmological N-body simulations
• Benchmarks and Explanations for Deep Learning Estimates of X-ray Galaxy Cluster Masses
• High-fidelity reproduction of central galaxy joint distributions with Neural Networks
• CHEX-MATE: A non-parametric deep learning technique to deproject and deconvolve galaxy cluster X-ray temperature profiles
• The Three Hundred Project: Mapping The Matter Distribution in Galaxy Clusters Via Deep Learning from Multiview Simulated Observations
• Identifying Galaxy Cluster Mergers with Deep Neural Networks using Idealized Compton-y and X-ray maps
• Bayesian and Convolutional Networks for Hierarchical Morphological Classification of Galaxies
• kSZ Pairwise Velocity Reconstruction with Machine Learning
• Cluster optical depth and pairwise velocity estimation using machine learning
• Deriving accurate galaxy cluster masses using X-ray thermodynamic profiles and graph neural networks
• Trustworthy scientific inference with generative models
• CosmoGLINT: Cosmological Generative Model for Line Intensity Mapping with Transformer
• Estimating Cluster Masses from SDSS Multi-band Images with Transfer Learning
• Estimating the triaxiality of massive clusters from 2D observables in MillenniumTNG with machine learning
• Generating Galaxy Clusters Mass Density Maps from Mock Multiview Images via Deep Learning