skip to main content

Title: GaMPEN: A Machine-learning Framework for Estimating Bayesian Posteriors of Galaxy Morphological Parameters

We introduce a novel machine-learning framework for estimating the Bayesian posteriors of morphological parameters for arbitrarily large numbers of galaxies. The Galaxy Morphology Posterior Estimation Network (GaMPEN) estimates values and uncertainties for a galaxy’s bulge-to-total-light ratio (LB/LT), effective radius (Re), and flux (F). To estimate posteriors, GaMPEN uses the Monte Carlo Dropout technique and incorporates the full covariance matrix between the output parameters in its loss function. GaMPEN also uses a spatial transformer network (STN) to automatically crop input galaxy frames to an optimal size before determining their morphology. This will allow it to be applied to new data without prior knowledge of galaxy size. Training and testing GaMPEN on galaxies simulated to matchz< 0.25 galaxies in Hyper Suprime-Cam Wideg-band images, we demonstrate that GaMPEN achieves typical errors of 0.1 inLB/LT, 0.″17 (∼7%) inRe, and 6.3 × 104nJy (∼1%) inF. GaMPEN's predicted uncertainties are well calibrated and accurate (<5% deviation)—for regions of the parameter space with high residuals, GaMPEN correctly predicts correspondingly large uncertainties. We also demonstrate that we can apply categorical labels (i.e., classifications such ashighly bulge dominated) to predictions in regions with high residuals and verify that those labels are ≳97% accurate. To the best of our knowledge, GaMPEN is the first machine-learning framework for determining joint posterior distributions of multiple morphological parameters and is also the first application of an STN to optical imaging in astronomy.

more » « less
Author(s) / Creator(s):
; ; ; ; ; ; ; ; ; ; ; ; ;
Publisher / Repository:
DOI PREFIX: 10.3847
Date Published:
Journal Name:
The Astrophysical Journal
Medium: X Size: Article No. 138
Article No. 138
Sponsoring Org:
National Science Foundation
More Like this
  1. Abstract

    We use the Galaxy Morphology Posterior Estimation Network (GaMPEN) to estimate morphological parameters and associated uncertainties for ∼8 million galaxies in the Hyper Suprime-Cam Wide survey withz≤ 0.75 andm≤ 23. GaMPEN is a machine-learning framework that estimates Bayesian posteriors for a galaxy’s bulge-to-total light ratio (LB/LT), effective radius (Re), and flux (F). By first training on simulations of galaxies and then applying transfer learning using real data, we trained GaMPEN with <1% of our data set. This two-step process will be critical for applying machine-learning algorithms to future large imaging surveys, such as the Rubin-Legacy Survey of Space and Time, the Nancy Grace Roman Space Telescope, and Euclid. By comparing our results to those obtained using light profile fitting, we demonstrate that GaMPEN’s predicted posterior distributions are well calibrated (≲5% deviation) and accurate. This represents a significant improvement over light profile fitting algorithms, which underestimate uncertainties by as much as ∼60%. For an overlapping subsample, we also compare the derived morphological parameters with values in two external catalogs and find that the results agree within the limits of uncertainties predicted by GaMPEN. This step also permits us to define an empirical relationship between the Sérsic index andLB/LTthat can be used to convert between these two parameters. The catalog presented here represents a significant improvement in size (∼10×), depth (∼4 mag), and uncertainty quantification over previous state-of-the-art bulge+disk decomposition catalogs. With this work, we also release GaMPEN’s source code and trained models, which can be adapted to other data sets.

    more » « less

    We use Bayesian convolutional neural networks and a novel generative model of Galaxy Zoo volunteer responses to infer posteriors for the visual morphology of galaxies. Bayesian CNN can learn from galaxy images with uncertain labels and then, for previously unlabelled galaxies, predict the probability of each possible label. Our posteriors are well-calibrated (e.g. for predicting bars, we achieve coverage errors of 11.8 per cent within a vote fraction deviation of 0.2) and hence are reliable for practical use. Further, using our posteriors, we apply the active learning strategy BALD to request volunteer responses for the subset of galaxies which, if labelled, would be most informative for training our network. We show that training our Bayesian CNNs using active learning requires up to 35–60 per cent fewer labelled galaxies, depending on the morphological feature being classified. By combining human and machine intelligence, Galaxy zoo will be able to classify surveys of any conceivable scale on a time-scale of weeks, providing massive and detailed morphology catalogues to support research into galaxy evolution.

    more » « less

    We study the alignments of galaxy spin axes with respect to cosmic web filaments as a function of various properties of the galaxies and their constituent bulges and discs. We exploit the SAMI Galaxy Survey to identify 3D spin axes from spatially resolved stellar kinematics and to decompose the galaxy into the kinematic bulge and disc components. The GAMA survey is used to reconstruct the cosmic filaments. The mass of the bulge, defined as the product of stellar mass and bulge-to-total flux ratio Mbulge = M⋆ × (B/T), is the primary parameter of correlation with spin–filament alignments: galaxies with lower bulge masses tend to have their spins parallel to the closest filament, while galaxies with higher bulge masses are more perpendicularly aligned. M⋆ and B/T separately show correlations, but they do not fully unravel spin–filament alignments. Other galaxy properties, such as visual morphology, stellar age, star formation activity, kinematic parameters, and local environment, are secondary tracers. Focussing on S0 galaxies, we find preferentially perpendicular alignments, with the signal dominated by high-mass S0 galaxies. Studying bulge and disc spin–filament alignments separately reveals additional information about the formation pathways of the corresponding galaxies: bulges tend to have more perpendicular alignments, while discs show different tendencies according to their kinematic features and the mass of the associated bulge. The observed correlation between the flipping of spin–filament alignments and the growth of the bulge can be explained by mergers, which drive both alignment flips and bulge formation.

    more » « less
  4. Abstract

    We measure optical colors for the bulges of 312 disk galaxies from the Carnegie-Irvine Galaxy Survey and convert their previously availableR-band structural parameters to stellar-mass parameters. We also measure their average stellar-mass surface density in the central 1 kpc (Σ1). Comparing the mass-based Kormendy relation with the original one based on flux, we find that the majority of the classifications into classical and pseudo bulges, as well as their overall statistical properties, remain essentially unchanged. While the bulge-type classifications of the Kormendy relation are robust against stellar population effects, the mass-based classification criteria do produce better agreement between bulge structural properties and their stellar populations. Moreover, the mass-based Kormendy relation reveals a population of ultradense bulges akin to high-zcompact early-type galaxies, which are otherwise hidden in the original Kormendy relation. These bulges are probably relics of spheroids assembled in the early universe, although for some we cannot rule out some contribution from secular growth. We confirm previous studies that Σ1correlates well with bulge surface densities.

    more » « less
  5. ABSTRACT We present Galaxy Zoo DECaLS: detailed visual morphological classifications for Dark Energy Camera Legacy Survey images of galaxies within the SDSS DR8 footprint. Deeper DECaLS images (r = 23.6 versus r = 22.2 from SDSS) reveal spiral arms, weak bars, and tidal features not previously visible in SDSS imaging. To best exploit the greater depth of DECaLS images, volunteers select from a new set of answers designed to improve our sensitivity to mergers and bars. Galaxy Zoo volunteers provide 7.5 million individual classifications over 314 000 galaxies. 140 000 galaxies receive at least 30 classifications, sufficient to accurately measure detailed morphology like bars, and the remainder receive approximately 5. All classifications are used to train an ensemble of Bayesian convolutional neural networks (a state-of-the-art deep learning method) to predict posteriors for the detailed morphology of all 314 000 galaxies. We use active learning to focus our volunteer effort on the galaxies which, if labelled, would be most informative for training our ensemble. When measured against confident volunteer classifications, the trained networks are approximately 99 per cent accurate on every question. Morphology is a fundamental feature of every galaxy; our human and machine classifications are an accurate and detailed resource for understanding how galaxies evolve. 
    more » « less