Skip to yearly menu bar Skip to main content


Oral
in
Workshop: Machine Learning for Astrophysics

Galaxy Merger Reconstruction with Equivariant Graph Normalizing Flows

Kwok Sun Tang · Yuan-Sen Ting


Abstract:

A key yet unresolved question in modern-day astronomy is how galaxies formed and evolved under the paradigm of the ΛCDM model. A critical limiting factor lies in the lack of robust tools to describe the merger history through a statistical model. In this work, we employ a generative graph network, E(n) Equivariant Graph Normalizing Flows Model. We demonstrate that, by treating the progenitors as a graph, our model robustly recovers their distributions, including their masses, merging redshifts and pairwise distances at redshift z = 2 conditioned on their z = 0 properties. The generative nature of the model enables other downstream tasks, including likelihood-free inference, detecting anomalies and identifying subtle correlations of progenitor features.

Chat is not available.