Authors: Ćiprijanović, Aleksandra 
Snyder, Gregory
Nord, Brian
Peek, Joshua
Affiliations: Mathematical Institute of the Serbian Academy of Sciences and Arts 
Title: DeepMerge: Classifying high-redshift merging galaxies with deep neural networks
Journal: Astronomy and Computing
Volume: 32
First page: 100390
Issue Date: 1-Jul-2020
Rank: M22
ISSN: 2213-1337
DOI: 10.1016/j.ascom.2020.100390
Abstract: 
We investigate and demonstrate the use of convolutional neural networks (CNNs) for the task of distinguishing between merging and non-merging galaxies in simulated images, and for the first time at high redshifts (i.e., z=2). We extract images of merging and non-merging galaxies from the Illustris-1 cosmological simulation and apply observational and experimental noise that mimics that from the Hubble Space Telescope; the data without noise form a “pristine” data set and that with noise form a “noisy” data set. The test set classification accuracy of the CNN is 79% for pristine and 76% for noisy. The CNN outperforms a Random Forest classifier, which was shown to be superior to conventional one- or two-dimensional statistical methods (Concentration, Asymmetry, the Gini, M20 statistics etc.), which are commonly used when classifying merging galaxies. We also investigate the selection effects of the classifier with respect to merger state and star formation rate, finding no bias. Finally, we extract Grad-CAMs (Gradient-weighted Class Activation Mapping) from the results to further assess and interrogate the fidelity of the classification model.
Keywords: Convolutional neural networks | Cosmology | Deep learning | Merging galaxies
Publisher: Elsevier
Project: Emission nebula: structure and evolution 
U.S. Department of Energy, Office of Science, Office of High Energy Physics, Contract No. DE-AC02-07CH11359
HST AR-Theory grant, program number 13887

Show full item record

SCOPUSTM   
Citations

28
checked on Jun 15, 2024

Page view(s)

42
checked on May 9, 2024

Google ScholarTM

Check

Altmetric

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.