A Graph Attention Network Framework for Simultaneous Prediction of Galaxy and Dark Matter Halo Properties in GAMA DR4
Abstract
We develop a graph-based deep learning model, GAMANet, to predict galaxy and halo properties using the GAMA DR4 dataset. Multiple catalogues are combined to construct a unified sample of 19,469 galaxies and derive a set of photometric, spectroscopic, morphological, and environmental features. The galaxy sample is represented as a graph, with galaxies as nodes and physically motivated connections based on spatial proximity and group membership as edges. A Graph Attention Network with GATv2 layers is then used to learn information from the galaxy neighbourhood while assigning different weights to neighbouring galaxies. The model simultaneously predicts halo mass, stellar mass, and a specific star formation rate proxy. On the test set, GAMANet achieves R2 = 0.958, $0.985$, and $0.982$ for halo mass, stellar mass, and SSFR proxy, respectively, with an RMSE of $0.237$ dex for halo mass. These results demonstrate that graph-based learning can extract useful information about galaxy and halo properties directly from observational data while incorporating the environmental relationships encoded in the galaxy graph.
Related
- ARENA: Active Region attENtion Aggregation for interpretable solar flare prediction
- TransitNet: A Compact Attention-Augmented Deep Learning Framework for Low-SNR Transit Blind Searches
- Data-driven prediction of atomic-oxygen erosion yield in polymers from bulk physical properties
- Solar Flare Prediction Using a Hybrid Convolutional Neural Network and Transformer Model
- Through a glass, darkly: a combined framework for estimating fast radio burst host galaxy and population properties in an era of uncertain host identification
BibTeX
@misc{aashish2026graph,
title = {{A Graph Attention Network Framework for Simultaneous Prediction of Galaxy and Dark Matter Halo Properties in GAMA DR4}},
author = {Sandeep Aashish and Abhay V Hegde and Aditya Prakash Gupta and Ashmit Verma and Md Saif and Nishanth Kumaran and Shivang Rayat},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.07836},
eprint = {2610.07836},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.07836},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - A Graph Attention Network Framework for Simultaneous Prediction of Galaxy and Dark Matter Halo Properties in GAMA DR4
AU - Sandeep Aashish
AU - Abhay V Hegde
AU - Aditya Prakash Gupta
AU - Ashmit Verma
AU - Md Saif
AU - Nishanth Kumaran
AU - Shivang Rayat
PY - 2026
DA - 2026/10/06
PB - arXiv
DO - 10.48550/arxiv.2610.07836
UR - https://arxiv.org/abs/2610.07836
AB - We develop a graph-based deep learning model, GAMANet, to predict galaxy and halo properties using the GAMA DR4 dataset. Multiple catalogues are combined to construct a unified sample of 19,469 galaxies and derive a set of photometric, spectroscopic, morphological, and environmental features. The galaxy sample is represented as a graph, with galaxies as nodes and physically motivated connections based on spatial proximity and group membership as edges. A Graph Attention Network with GATv2 layers is then used to learn information from the galaxy neighbourhood while assigning different weights to neighbouring galaxies. The model simultaneously predicts halo mass, stellar mass, and a specific star formation rate proxy. On the test set, GAMANet achieves R^2 = 0.958, $0.985$, and $0.982$ for halo mass, stellar mass, and SSFR proxy, respectively, with an RMSE of $0.237$ dex for halo mass. These results demonstrate that graph-based learning can extract useful information about galaxy and halo properties directly from observational data while incorporating the environmental relationships encoded in the galaxy graph.
ER -