Dynamic Diffusion Schr\"odinger Bridge in Astrophysical Observational Inversions

TL;DR

This paper introduces Astro-DSB, a diffusion Schr"odinger Bridge model tailored for astrophysical inverse problems, demonstrating improved interpretability, efficiency, and predictive accuracy in both simulated and real astrophysical data, advancing physics-aware generative modeling.

Contribution

The paper presents Astro-DSB, a novel diffusion Schr"odinger Bridge variant designed for astrophysical dynamics, enhancing interpretability and performance in observational inverse prediction tasks.

Findings

Improved interpretability and efficiency over traditional methods.

Enhanced prediction accuracy in real and simulated astrophysical data.

Better generalization in Out-Of-Distribution tests with unseen physical conditions.

Abstract

We study Diffusion Schr\"odinger Bridge (DSB) models in the context of dynamical astrophysical systems, specifically tackling observational inverse prediction tasks within Giant Molecular Clouds (GMCs) for star formation. We introduce the Astro-DSB model, a variant of DSB with the pairwise domain assumption tailored for astrophysical dynamics. By investigating its learning process and prediction performance in both physically simulated data and in real observations (the Taurus B213 data), we present two main takeaways. First, from the astrophysical perspective, our proposed paired DSB method improves interpretability, learning efficiency, and prediction performance over conventional astrostatistical and other machine learning methods. Second, from the generative modeling perspective, probabilistic generative modeling reveals improvements over discriminative pixel-to-pixel modeling in Out-Of-Distribution (OOD) testing cases of physical simulations with unseen initial conditions and different dominant physical processes. Our study expands research into diffusion models beyond the traditional visual synthesis application and provides evidence of the models' learning abilities beyond pure data statistics, paving a path for future physics-aware generative models which can align dynamics between machine learning and real (astro)physical systems.