Astronomy & Space

AI Maps Hidden Stellar Streams in the Galaxy’s Outskirts

How the science connects

Machine learning

AI Insight

Researchers developed a machine learning approach using contrastive learning (NNCLR algorithm) to identify and characterize faint tidal features around galaxies in Dark Energy Survey data. They trained a neural network on 38,334 galaxy images using a novel image scaling technique that suppresses bright central galaxy features to better detect low surface brightness structures. While the method successfully identified large-scale merger features, it required additional refinement to reliably detect stellar streams, demonstrating both promise and current limitations for automated detection of faint galactic structures.


This work provides a framework for automatically detecting rare tidal features in massive astronomical datasets from current and future sky surveys, which would be impractical to identify through manual inspection alone. The preprocessing techniques developed here could improve detection of faint structures that reveal galaxy formation and interaction histories.


⚠️ Preprint – Noch nicht peer-reviewed

Dieser Artikel wurde noch nicht von unabhängigen Experten begutachtet. Die Ergebnisse sind vorläufig und sollten mit Vorsicht interpretiert werden.

Abstract: We present a self-supervised approach for characterizing low surface brightness tidal features in wide-field imaging data by applying the nearest-neighbor contrastive learning of visual representations (NNCLR) algorithm to a curated subset of the Dark Energy Survey Data Release 2 (DES DR2). We construct 38,334 cutouts of well-resolved galaxies in the g, r, i bands, applying a novel “tiered sigmoid scaling function” to dynamically adjust image contrast according to the object’s signal-to-noise and background level. A supplemental labeled sample of 366 galaxies enables qualitative assessment of the learned embeddings. We train a convolutional neural network with image augmentations including injection of simulated background stars, and project the resulting 512-dimensional representations into two dimensions using uniform manifold approximation and projection (UMAP) and its local density preserving variant (densMAP). We find that the NNCLR latent space recovers global trends corresponding to major merger features, yet does not reliably separate stellar streams without further supervision. To interpret the network’s implicit attention, we compute gradient-based saliency maps averaged over the full dataset: these reveal that the tiered sigmoid scaling effectively attenuates information from the center of the image cutouts, thereby suppressing the learning of high surface brightness features of each image cutout’s central galaxy. Our study provides a blueprint for leveraging contrastive methods to mine forthcoming survey data for faint tidal substructure, and highlights key preprocessing and interpretability considerations for robust stream detection.

Source: Contrastive learning of extragalactic stellar streams. Sculpting a latent space of representations with DES DR2 photometry