Presentation #110.03 in the session “Stellar/Compact (Poster)”.
The X-ray sky is rich with transient and variable sources many of which are serendipitously discovered with Chandra, XMM-Newton, and eROSITA. Machine Learning (ML) algorithms allow one to quickly explore the astrophysical nature of thousands of transient and/or variable sources. The multi-wavelength ML classification pipeline (MUWCLASS), developed by our group, uses the supervised ML random forest algorithm. It takes into account the measurement uncertainties and provides probabilities for each classified source to belong for each of the classes in the training dataset. The latter includes reliably classified, published sources with X-ray counterparts from Chandra Source Catalog version 2.0 (CSCv2) and multi-wavelength properties from multiple all-sky surveys. We demonstrate the pipeline capabilities by classifying Galactic variable sources from CSCv2 and discuss possible astrophysical implications. We envision a wide range of potential applications of the MUWCLASS pipeline by the high energy astrophysics community, including the classifications of large numbers of serendipitously detected X-ray sources for population studies and searches for rare exotic sources.