Galaxy Imaging with Generative Models: Insights from a Two-Models Framework

Abstract

Generative models have recently revolutionized image generation tasks across diverse domains, including galaxy image synthesis. This study investigates the statistical learning and consistency of three generative models: light-weight-gan (a GAN-based model), Glow (a Normalizing Flow-based model), and a diffusion model based on a U-Net denoiser, all trained on non-overlapping subsets of the SDSS dataset of 64x64 grayscale images. While all models produce visually realistic images with well-preserved morphological variable distributions, we focus on their ability to learn and generalize the underlying data distribution. The diffusion model shows a transition from memorization to generalization as the dataset size increases, confirming previous findings. Smaller datasets lead to overfitting, while larger datasets enable novel sample generation, supported by the denoising process's theoretical basis. For the flow-based model, we propose an "inversion test" leveraging its bijective nature. Similarly, the GAN-based model achieves comparable morphological consistency but lacks bijectivity. We then introduce a "discriminator test", which shows successful learning for larger datasets but poorer confidence with smaller ones. Across all models, dataset sizes below O(100,000) pose challenges to learning. Along our experiments, the "two-models" framework enables robust evaluations, highlighting both the potential and limitations of these models. These findings provide valuable insights into statistical learning in generative modeling, with applications certainly extending beyond galaxy image generation.

16 pages, 16 figures, to be published in MNRAS

References in corpus (22)

• SciPy 1.0--Fundamental Algorithms for Scientific Computing in Python
• The Sloan Digital Sky Survey: Technical Summary
• The Seventh Data Release of the Sloan Digital Sky Survey
• Hierarchical Text-Conditional Image Generation with CLIP Latents
• LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loop
• The Relationship Between Stellar Light Distributions of Galaxies and their Formation Histories
• A New Non-Parametric Approach to Galaxy Morphological Classification
• The optical morphologies of galaxies in the IllustrisTNG simulation: a comparison to Pan-STARRS observations
• Structural and Photometric Classification of Galaxies - I. Calibration Based on a Nearby Galaxy Sample
• The Creativity of Text-to-Image Generation
• Generative Adversarial Networks recover features in astrophysical images of galaxies beyond the deconvolution limit
• Galaxy Morphology and Star Formation in the Illustris Simulation at z=0
• Geometric GAN
• Evaluating generative models in high energy physics
• Deep Generative Models for Galaxy Image Simulations
• Deblending galaxies with Variational Autoencoders: a joint multi-band, multi-instrument approach
• On the Design Fundamentals of Diffusion Models: A Survey
• Forging new worlds: high-resolution synthetic galaxies with chained generative adversarial networks
• Probabilistic Mass Mapping with Neural Score Estimation
• Probabilistic Forward Modeling of Galaxy Catalogs with Normalizing Flows
• Deblending Galaxies with Generative Adversarial Networks
• Hierarchic Flows to Estimate and Sample High-dimensional Probabilities

Cited by in corpus (2)

• Deep learning-based astronomical multimodal data fusion: A comprehensive review
• Can AI Dream of Unseen Galaxies? Conditional Diffusion Model for Galaxy Morphology Augmentation