Assessing Galaxy Rotation Kinematics: Insights from Convolutional Neural Networks on Velocity Variations

Abstract

Distinguishing galaxies as either fast or slow rotators plays a vital role in understanding the processes behind galaxy formation and evolution. Standard techniques, which are based on the ${\lambda }_R$-spin parameter obtained from stellar kinematics, frequently face difficulties to classify fast and slow rotators accurately. These challenges arise particularly in cases where galaxies have complex interaction histories or exhibit significant morphological diversity. In this paper, we evaluate the performance of a Convolutional Neural Network (CNN) on classifying galaxy rotation kinematics based on stellar kinematic maps from the SAMI survey. Our results show that the optimal CNN architecture achieves an accuracy and precision of approximately 91% and 95% on the test dataset, respectively. Subsequently, we apply our trained model to classify previously unknown rotator galaxies for which traditional statistical tools have been unable to determine whether they exhibit fast or slow rotation, such as certain irregular galaxies or those in dense clusters. We also used Integrated Gradients (IG) to reveal the crucial kinematic features that influenced the CNN's classifications. This research highlights the power of CNNs to improve our comprehension of galaxy dynamics and emphasizes their potential to contribute to upcoming large-scale Integral Field Spectrograph (IFS) surveys.

18 pages, 10 figures. Accepted to the Special Issue "Universe: Feature Papers 2024: Galaxies and Clusters" of MDPI Galaxies

References in corpus (41)

• The SAURON project -- IX. A kinematic classification for early-type galaxies
• The ATLAS3D project - III. A census of the stellar angular momentum within the effective radius of early-type galaxies: unveiling the distribution of Fast and Slow Rotators
• The SAURON project -- X. The orbital anisotropy of elliptical and lenticular galaxies: revisiting the (V/sigma,epsilon) diagram with integral-field stellar kinematics
• The Sydney-AAO Multi-object Integral field spectrograph (SAMI)
• Physical properties and environments of nearby galaxies
• The SAMI Galaxy Survey: instrument specification and target selection
• New Approaches for Calculating Moran's Index of Spatial Autocorrelation
• The origin of angular momentum in dark matter halos
• The ATLAS${}^{3D}$ project - XXV: Two-dimensional kinematic analysis of simulated galaxies and the cosmological origin of fast and slow rotators
• Cross-Entropy Loss Functions: Theoretical Analysis and Applications
• Performance of AAOmega: the AAT multi-purpose fibre-fed spectrograph
• Enhancing Gravitational-Wave Science with Machine Learning
• The SAMI Galaxy Survey: Revisiting Galaxy Classification Through High-Order Stellar Kinematics
• The SAMI Galaxy Survey: Early Data Release
• Rotation and anisotropy of galaxies revisited
• Machine-learning non-stationary noise out of gravitational wave detectors
• The SAMI Galaxy Survey: Cubism and covariance, putting round pegs into square holes
• A deep learning approach to cosmological dark energy models
• Image-based deep learning for classification of noise transients in gravitational wave detectors
• The SAMI Galaxy Survey: Data Release Two with absorption-line physics value-added products
• Identifying Strong Lenses with Unsupervised Machine Learning using Convolutional Autoencoder
• The SAMI Galaxy Survey: Data Release One with Emission-line Physics Value-Added Products
• ExoMiner: A Highly Accurate and Explainable Deep Learning Classifier that Validates 301 New Exoplanets
• SAURON's Challenge for the Major Merger Scenario of Elliptical Galaxy Formation
• Interpreting deep learning models for weak lensing
• An interpretable machine learning framework for dark matter halo formation
• Ultimate age-dating method for galaxy groups; clues from the Millennium Simulations
• Deep Learning for Strong Lensing Search: Tests of the Convolutional Neural Networks and New Candidates from KiDS DR3
• Convolutional Neural Networks for Direct Detection of Dark Matter
• Detecting gravitational lenses using machine learning: exploring interpretability and sensitivity to rare lensing configurations
• Evolution of galaxy groups in the Illustris simulation
• A numerical twist on the observational spin parameter, ${\lambda }_R$
• The origin of spin in galaxies: clues from simulations of atomic cooling halos
• A semi-supervised approach to dark matter searches in direct detection data with machine learning
• The impact of the dynamical state of galaxy groups on the stellar populations of central galaxies
• Merger history of central galaxies in Semi-Analytic Models of galaxy formation
• The SAMI Galaxy Survey: Kinematics of stars and gas in brightest group galaxies; the role of group dynamics
• Hydrodynamic simulations of the Disk of Gas Around Supermassive black holes (HDGAS) -I; Molecular Gas Dynamics
• Distinguishing Coupled Dark Energy Models with Neural Networks
• Clusternets: A deep learning approach to probe clustering dark energy
• Reducing false positives in strong lens detection through effective augmentation and ensemble learning