Intrinsic Dimension Estimation for Radio Galaxy Zoo using Diffusion Models

TL;DR

This paper estimates the intrinsic dimension of the Radio Galaxy Zoo dataset using diffusion models, revealing higher complexity in out-of-distribution sources and variations across morphological classes and SNR levels.

Contribution

It introduces a novel approach to estimate intrinsic dimension in radio astronomy data using score-based diffusion models and analyzes its variation across different source types and noise levels.

Findings

Out-of-distribution sources have higher intrinsic dimension.

RGZ dataset's intrinsic dimension exceeds that of natural images.

Weak correlation between SNR and intrinsic dimension.

Abstract

In this work, we estimate the intrinsic dimension (iD) of the Radio Galaxy Zoo (RGZ) dataset using a score-based diffusion model. We examine how the iD estimates vary as a function of Bayesian neural network (BNN) energy scores, which measure how similar the radio sources are to the MiraBest subset of the RGZ dataset. We find that out-of-distribution sources exhibit higher iD values, and that the overall iD for RGZ exceeds those typically reported for natural image datasets. Furthermore, we analyse how iD varies across Fanaroff-Riley (FR) morphological classes and as a function of the signal-to-noise ratio (SNR). While no relationship is found between FR I and FR II classes, a weak trend toward higher SNR at lower iD. Future work using the RGZ dataset could make use of the relationship between iD and energy scores to quantitatively study and improve the representations learned by various self-supervised learning algorithms.