Featured

Learning the Cosmic Web: Graph-based Classification of Simulated Galaxies by their Dark Matter Environments

Authors: Dakshesh Kololgi, Krishna Naidoo, Amelie Saintonge, Ofer Lahav
Status: Submitted (astro-ph.GA), 2025

• arXiv: arXiv:2512.05909
• DOI: 10.48550/arXiv.2512.05909
• Key result: 85% accuracy for galaxy cosmic-web environment classification using graph-attention learning on Delaunay-derived metrics.

Abstract snapshot

This paper introduces a graph-based framework to infer the dark matter cosmic web environment of galaxies from simulation data. Galaxies are first assigned T-Web labels from the matter field, then encoded with Delaunay graph metrics, and finally classified with a graph attention network. The method outperforms graph-agnostic baselines and supports future survey applications.

Full Publication List

Add upcoming publications, conference papers, and preprints here as they are released.