Exploring the Universe with SNAD: Anomaly Detection in Astronomy

TL;DR

This paper provides a review of the SNAD project and summarizes the advancements and achievements made by the team over several years.

Abstract

SNAD is an international project with a primary focus on detecting astronomical anomalies within large-scale surveys, using active learning and other machine learning algorithms. The work carried out by SNAD not only contributes to the discovery and classification of various astronomical phenomena but also enhances our understanding and implementation of machine learning techniques within the field of astrophysics. This paper provides a review of the SNAD project and summarizes the advancements and achievements made by the team over several years.

Figures (4)

• Figure 1: Fit of SNAD160 using equation \ref{['eq:slsn']} resulting from MvSR procedure. The magnitude is normalized by subtraction of the peak magnitude.
• Figure 2: Two examples of red dwarf flares candidates found by SNAD team.
• Figure 3: Some examples from the SNAD catalog of artefacts.
• Figure 4: SNAD Miner schematic. We use bright ZTF SNe simulations (left) and extract their LC features (left center). Then, we apply a (PCA+) k-D tree on these features, to search for real ZTF DR events nearest neighbors (right center). Some of these matched nearest neighbors were previously missed SNe (right).