Characterization of Residual Morphological Substructure Using Supervised and Unsupervised Deep Learning
Department of Physics & Astronomy, University of Missouri Kansas City（物理与天文学系，密苏里大学堪萨斯城分校）；Space Telescope Science Institute（空间望远镜科学研究所）；Department of Physics, University of California, Santa Cruz（物理系，加州大学圣克鲁兹分校）；Department of Physics and Astronomy, University of Missouri（物理与天文学系，密苏里大学）
Kameswara Bharadwaj Mantha, Daniel H. McIntosh, Cody Ciaschi, Rubyet Evan, Luther Landry, Henry C. Ferguson, Camilla Pacifici, Joel Primack, Nimish Hathi, Anton… 展开作者
Kameswara Bharadwaj Mantha, Daniel H. McIntosh, Cody Ciaschi, Rubyet Evan, Luther Landry, Henry C. Ferguson, Camilla Pacifici, Joel Primack, Nimish Hathi, Anton Koekemoer, Yicheng Guo, The CANDELS Collaboration
备注：
This manuscript is a preprint that has not undergone peer review and is being shared to ensure dissemination and community access to the results and insights (see acknowledgements)
Automated characterization of galactic substructure is an essential step in understanding the transformative physical processes driving galaxy evolution. In this study, we investigate the application of deep learning (DL) frameworks to characterize different galactic substructures hosted within parametric light-profile subtracted ``residual'' images of a large sample galaxies from the CANDELS survey. We develop a supervised Convolutional Neural Network (CNN) and unsupervised Convolutional Variational Autoencoder (CvAE) and train it on the single-Sérsic profile fitting based residual images of $10,046$ bright and massive galaxies ($H<24.5\,{\rm mag}$ and $M_{\rm stellar} \geq 10^{9.5}\,M_{\odot}$) spanning $1<z<3$, in conjunction with their visual-based classification labels indicating the nature of residual substructures hosted within them. Using our unique data preprocessing approach, we prepare our residual images such that the inputs to our DL networks comprise only ``galaxy of interest'', and augment them such that our sample span uniformly across different residual characteristics. We assess the latent space of the CNN and CvAE using Principle Component Analysis (PCA) along with independently quantified metrics of residual strength (significant pixel flux $SPF$, Bumpiness, and Residual Flux Fraction). We also employ an unsupervised Gaussian Mixture Modeling (GMM) based clustering scheme with Support Vector Classification (SVC) to identify groupings in PCA space that correspond to similar residual substructure. We find that our supervised CNN latent features in PCA space correlate with the $SPF$ values and distinguish between qualitatively strong and weak residual substructures. While our unsupervised CvAE latent space also correlates with visual and quantitative residual characteristics, but lacks clear discriminatory power when characterizing different residual substructures.
Robustness Analysis of USmorph: I. Generalization Efficiency of Unsupervised Strategies and Supervised Learning in Galaxy Morphological Classification
Anqing Normal University（安庆师范大学）；Institute of Astronomy and Astrophysics, Anqing Normal University（安庆师范大学天文与物理研究所）；Shanghai Astronomical Observatory, Chinese Academy of Sciences（中国科学院上海天文台）；University of Chinese Academy of Sciences（中国科学院大学）；Dali University（大理大学）；University of Science and Technology of China（中国科学技术大学）；Institute of Deep Space Sciences, Deep Space Exploration Laboratory（深空探测实验室深空科学研究所）
Unsupervised Chemo-Dynamical Dissection of the Inner Galactic Halo: Discovery of Five Accreted Substructures with SDSS-V and Gaia
Institute of Graduate Studies in Science, Istanbul University（伊斯坦布尔大学研究生科学学院）；Faculty of Science, Department of Astronomy and Space Sciences, Istanbul University（伊斯坦布尔大学理学院天文与空间科学系）
Star formation and morphological trends in the Antlia cluster: Probing environmental influence out to 5R200
Universidad de La Serena（拉塞雷纳大学）；Instituto de Astronomia, Geofísica e Ciências Atmosféricas da Universidade de São Paulo（圣保罗大学天文、地球物理与大气科学研究所）；Facultad de Ciencias Astronómicas y Geofísicas, Universidad Nacional de La Plata（拉普拉塔国立大学天文与地球物理科学学院）；Universidad de Talca（塔尔卡大学）；Millennium Nucleus for Galaxies (MINGAL)（银河系千年核心（MINGAL））；Instituto de Astrofísica de La Plata, CONICET-UNLP（拉普拉塔天体物理研究所，阿根廷国家科学研究委员会-拉普拉塔国立大学）
Robustness Analysis of USmorph: II. Optimizing Feature Extraction, Dimensionality Reduction, and Clustering for Unsupervised Galaxy Morphology Classification
Anqing Normal University（安庆师范大学）；Zhengzhou University（郑州大学）；University of Science and Technology of China（中国科学技术大学）
Formation of Substructure in Luminous Submillimeter Galaxies (FOSSILS): Initial sample and the discovery of a dusty spiral at cosmic noon
SOKENDAI (The Graduate University for Advanced Studies)（日本综合研究大学院大学）；National Astronomical Observatory of Japan（日本国立天文台）；Hokkai-Gakuen University（北海学园大学）
A robust morphological classification method for galaxies using dual-encoding contrastive learning and multi-clustering voting on JWST/NIRCam images
School of Mathematics and Physics, Anqing Normal University（安庆师范大学数学与物理学院）；Institute of Astronomy and Astrophysics, Anqing Normal University（安庆师范大学天文与天体物理研究所）；Key Laboratory of Modern Astronomy and Astrophysics (Nanjing University), Ministry of Education（教育部现代天文与天体物理重点实验室（南京大学））；Department of Physics, The Chinese University of Hong Kong（香港中文大学物理系）；Shanghai Astronomical Observatory, Chinese Academy of Sciences（中国科学院上海天文台）；School of Astronomy and Space Science, University of Chinese Academy of Sciences（中国科学院大学天文与空间科学学院）；School of Engineering, Dali University（大理大学工程学院）
