DeepMerge: Classifying High-redshift Merging Galaxies with Deep Neural Networks

TL;DR

This paper demonstrates that convolutional neural networks can effectively classify high-redshift merging galaxies in simulated images, outperforming traditional statistical methods and providing insights into model interpretability.

Contribution

The study introduces a CNN-based approach for high-redshift galaxy merger classification, including noise robustness and interpretability via Grad-CAMs, a novel application at such redshifts.

Findings

CNN achieves 79% accuracy on pristine data and 76% on noisy data.

CNN outperforms Random Forest and traditional statistical classifiers.

No bias found in classification with respect to merger state or star formation rate.

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, $M_{20}$ 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.