Using Galaxy Evolution as Source of Physics-Based Ground Truth for Generative Models

TL;DR

This paper explores using galaxy evolution data as a physics-based benchmark for generative models, demonstrating that physics-aware metrics can better evaluate model realism and performance.

Contribution

It introduces physics-motivated metrics to assess generative models of galaxy images and compares diffusion and variational autoencoder models using these metrics.

Findings

Both models produce realistic galaxies based on human judgment.

Physics-based metrics reveal differences in model performance.

DDPM outperforms CVAE on most physics-based metrics.

Abstract

Generative models producing images have enormous potential to advance discoveries across scientific fields and require metrics capable of quantifying the high dimensional output. We propose that astrophysics data, such as galaxy images, can test generative models with additional physics-motivated ground truths in addition to human judgment. For example, galaxies in the Universe form and change over billions of years, following physical laws and relationships that are both easy to characterize and difficult to encode in generative models. We build a conditional denoising diffusion probabilistic model (DDPM) and a conditional variational autoencoder (CVAE) and test their ability to generate realistic galaxies conditioned on their redshifts (galaxy ages). This is one of the first studies to probe these generative models using physically motivated metrics. We find that both models produce comparable realistic galaxies based on human evaluation, but our physics-based metrics are better able to discern the strengths and weaknesses of the generative models. Overall, the DDPM model performs better than the CVAE on the majority of the physics-based metrics. Ultimately, if we can show that generative models can learn the physics of galaxy evolution, they have the potential to unlock new astrophysical discoveries.