Comparing Active Anomaly Detection Frameworks: From Light Curves to Images
Abstract
The increasing volume and complexity of data from modern astronomical surveys are making comprehensive visual inspection increasingly impractical, motivating automated methods for identifying rare and unusual objects. Active anomaly detection (A-AD) combines unsupervised anomaly detection with iterative user feedback, allowing searches to adapt to scientific interests while retaining sensitivity to unexpected objects. We systematically compare three A-AD frameworks, Astronomaly, Protege, and PineForest, across Zwicky Transient Facility light curves and Euclid galaxy images. In baseline experiments, Protege, PineForest, and Astronomaly recovered 83, 78, and 38 variable stars, respectively, within the first 100 light-curve objects inspected, while PineForest recovered 70 merging or morphologically disturbed galaxies within the first 100 Euclid images, compared with 29 and 19 for Astronomaly and Protege. Experiments using prior labels show that their effectiveness depends on the composition of the prior sample and varies between A-AD methods. Finally, we combine the complementary behaviour of Protege and PineForest in a hybrid search of 10,000 Euclid images, recovering 799 morphologically unusual objects. These include 29 retained strong lens systems and candidates, of which 9 candidates have no counterparts in the published Euclid Q1 strong lens catalogues searched, and 202 visually classified ring galaxies. These results highlight the importance of selecting A-AD strategies according to both the scientific objective and dataset properties, and demonstrate their potential for efficient, discovery-driven exploration of forthcoming astronomical surveys.
Related
- New Frameworks for Constructing Special Curves via f-Weighted Direction Curves
- Comparing Optimized Systematic Error Correction Methods on Selected TESS Light Curves
- Single-Parameter Exponential Model Outperforms Standard Dark Matter and Modified Newtonian Frameworks in Predicting Galactic Rotation Curves
- Anomaly detection in Fink. I. Discovery, follow-up, and classification of unusual sources
- Comparing Infrared & X-ray-selected Active Galactic Nuclei host galaxy star formation rates to the galaxy main sequence
BibTeX
@misc{sreejith2026comparing,
title = {{Comparing Active Anomaly Detection Frameworks: From Light Curves to Images}},
author = {Sreevarsha Sreejith and Robert C. Nichol},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.36096},
eprint = {2609.36096},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.36096},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Comparing Active Anomaly Detection Frameworks: From Light Curves to Images
AU - Sreevarsha Sreejith
AU - Robert C. Nichol
PY - 2026
DA - 2026/09/28
PB - arXiv
DO - 10.48550/arxiv.2609.36096
UR - https://arxiv.org/abs/2609.36096
AB - The increasing volume and complexity of data from modern astronomical surveys are making comprehensive visual inspection increasingly impractical, motivating automated methods for identifying rare and unusual objects. Active anomaly detection (A-AD) combines unsupervised anomaly detection with iterative user feedback, allowing searches to adapt to scientific interests while retaining sensitivity to unexpected objects. We systematically compare three A-AD frameworks, Astronomaly, Protege, and PineForest, across Zwicky Transient Facility light curves and Euclid galaxy images. In baseline experiments, Protege, PineForest, and Astronomaly recovered 83, 78, and 38 variable stars, respectively, within the first 100 light-curve objects inspected, while PineForest recovered 70 merging or morphologically disturbed galaxies within the first 100 Euclid images, compared with 29 and 19 for Astronomaly and Protege. Experiments using prior labels show that their effectiveness depends on the composition of the prior sample and varies between A-AD methods. Finally, we combine the complementary behaviour of Protege and PineForest in a hybrid search of 10,000 Euclid images, recovering 799 morphologically unusual objects. These include 29 retained strong lens systems and candidates, of which 9 candidates have no counterparts in the published Euclid Q1 strong lens catalogues searched, and 202 visually classified ring galaxies. These results highlight the importance of selecting A-AD strategies according to both the scientific objective and dataset properties, and demonstrate their potential for efficient, discovery-driven exploration of forthcoming astronomical surveys.
ER -