International Conference on Machine Learning for Astrophysics 3rd Ed. - ML4ASTRO3

Session

Extragalactic Science

2 Sept 2026, 09:00

Aula Prima (University of Malta)

Conveners

Extragalactic Science: I

• Andrea DeMarco (University of Malta)

Extragalactic Science: II

• Andrea DeMarco (University of Malta)

Extragalactic Science: III

• Filomena Bufano (Istituto Nazionale di Astrofisica (INAF))

68. [INVITED] Machine Learning and Extragalactic Science

Nicola Rosario Napolitano

02/09/2026, 09:00

69. More Than the Sum of Its Parts - Multimodal Learning for Galaxy Evolution in JWST Pure Parallel Observations

Ivelina Momcheva

02/09/2026, 09:40

Deriving physical properties of galaxies requires high information density: deep multi-band imaging, spectroscopy, and/or photometric catalogs together. Deep fields provide this richness over small areas, but wide-field surveys cover larger volumes with far sparser data, creating a tension between area and information content.
We present a self-supervised multimodal framework that learns a...

70. Using UMAP Dimensionality Reduction to Map the Color-Redshift Relation

Finian Ashmead

02/09/2026, 10:00

The volume of galaxy imaging data continues to outpace that of the high quality spectroscopic data used to precisely measure physical parameters like redshift and specific star formation rate (sSFR). In the near future, this problem will be exacerbated as facilities like Rubin and Roman rapidly build up sky maps extending to higher redshifts and dimmer fluxes, regimes where the spectroscopic...

71. Learning AGN multimodal characterization through optical, x-ray and host galaxy properties

Noemi Lery Borrelli

02/09/2026, 10:20

Active Galactic Nuclei are powered by accretion onto supermassive black holes and exhibit a wide range of observed properties. While the Unified Model explains part of this diversity through orientation effects, growing observational evidence suggests that additional factors-such as evolutionary stage and host galaxy properties-also play a significant role.
In this context, we adopt a...

72. 4MOST/ByCycle: Detecting Cool Circumgalactic Medium Quasar Absorbers via Contextual Anomaly Detection with Variational Autoencoders

Nicolás Guerra-Varas

02/09/2026, 11:10

Even though the gas around and between galaxies (the circumgalactic and intergalactic media, CGM and IGM) harbour about 90% of the baryons in the Universe, they remain significantly less understood than the matter associated with starlight from galaxies. In this talk, I will introduce the Baryon Cycle (ByCycle) project, a large high-resolution (R~20,000) 4MOST/VISTA spectroscopic survey that...

73. Needles in the cosmic haystack: Innovating the search for Reionization-Era quasars with self-supervised machine learning

Laura Natalia Martínez Ramírez

02/09/2026, 11:30

Luminous quasars at the highest redshifts (z>6) are key laboratories for studying early supermassive black hole (SMBH) growth and the physical conditions of the Universe during cosmic reionization. However, their extremely low spatial density (< 1 per Gpc³), combined with severe contamination from foreground ultracool dwarfs that outnumber them by up to four orders of magnitude, makes their...

76. From Projection to Reconstruction: Predicting Peculiar Velocities and Internal Distances in Galaxy Clusters via Graph Neural Networks

Sheng-Chieh Lin

02/09/2026, 11:50

Redshift measurements of galaxies within galaxy clusters play a crucial role in cosmology and astrophysics, serving as an essential key for cluster identification and related cosmological studies. However, the phase space distribution of cluster galaxies is not yet fully understood, as the observable positions and velocities are subject to line-of-sight projection effects. In this...

77. End-to-end differentiable forward modeling and inverse inference for integral field spectroscopic data

Anna Lena Schaible

02/09/2026, 14:00

Modern integral field spectroscopic (IFS) observations provide unprecedented detail of galaxy dynamics, but directly comparing these integrated light observations to discrete particle outputs from cosmological simulations remains challenging. Traditional forward-modeling tools are often CPU-bound and lack derivative information, hindering their application in modern gradient-assisted machine...

78. J-PAS: A Neural Network Approach to Single Stellar Population Characterization (remote)

Helena Domínguez Sánchez

02/09/2026, 14:20

J-PAS (Javalambre Physics of the Accelerating Universe Astrophysical Survey) will present a groundbreaking photometric survey covering 8500 deg2 of the visible sky from Javalambre, capturing data in 56 narrow band filters. This survey promises to revolutionize galaxy evolution studies by observing ~10^8 galaxies with low spectral resolution. A crucial aspect of this analysis involves...

79. Deep learning approaches to galaxy merger identification and classification

Subhrata Dey

02/09/2026, 14:40

Galaxy mergers are fundamental to the hierarchical assembly and evolution of galaxies, often driving starburst activity and AGN fueling. Identifying mergers and their stages, such as pre- and post-coalescence, from imaging alone, especially given the vast size of modern datasets, remains extremely challenging. We develop a supervised deep learning framework using Convolutional Neural Networks...

80. TORRCH: Field-level Reconstruction of Reionization Morphology from Galaxy Tracers (remote)

Soumak Maitra

02/09/2026, 15:00

The spatial morphology of Reionization contains information that is largely compressed away by global measurements of the volume-averaged neutral fraction. Ionized bubbles, neutral islands, and their connection to the high-redshift galaxy distribution encode the topology of reionization and the nature of the ionizing sources. In this talk, I will present TORRCH (TOmographic Reconstruction of...