Self-Supervised Clustering on Image-Subtracted Data with Deep-Embedded Self-Organizing Map
Abstract
Developing an effective automatic classifier to separate genuine sources from artifacts is essential for transient follow-ups in wide-field optical surveys. The identification of transient detections from the subtraction artifacts after the image differencing process is a key step in such classifiers, known as real-bogus classification problem. We apply a self-supervised machine learning model, the deep-embedded self-organizing map (DESOM) to this "real-bogus" classification problem. DESOM combines an autoencoder and a self-organizing map to perform clustering in order to distinguish between real and bogus detections, based on their dimensionality-reduced representations. We use 32x32 normalized detection thumbnails as the input of DESOM. We demonstrate different model training approaches, and find that our best DESOM classifier shows a missed detection rate of 6.6% with a false positive rate of 1.5%. DESOM offers a more nuanced way to fine-tune the decision boundary identifying likely real detections when used in combination with other types of classifiers, for example built on neural networks or decision trees. We also discuss other potential usages of DESOM and its limitations.
Cite
@article{arxiv.2209.06375,
title = {Self-Supervised Clustering on Image-Subtracted Data with Deep-Embedded Self-Organizing Map},
author = {Y. -L. Mong and K. Ackley and T. L. Killestein and D. K. Galloway and M. Dyer and R. Cutter and M. J. I. Brown and J. Lyman and K. Ulaczyk and D. Steeghs and V. Dhillon and P. O'Brien and G. Ramsay and K. Noysena and R. Kotak and R. Breton and L. Nuttall and E. Palle and D. Pollacco and E. Thrane and S. Awiphan and U. Burhanudin and P. Chote and A. Chrimes and E. Daw and C. Duffy and R. Eyles-Ferris and B. P. Gompertz and T. Heikkila and P. Irawati and M. Kennedy and A. Levan and S. Littlefair and L. Makrygianni and T. Marsh and D. Mata Sanchez and S. Mattila and J. R. Maund and J. McCormac and D. Mkrtichian and J. Mullaney and E. Rol and U. Sawangwit and E. Stanway and R. Starling and P. Strom and S. Tooke and K. Wiersema},
journal= {arXiv preprint arXiv:2209.06375},
year = {2022}
}