Explainable machine learning classification of \textit{Chandra} X-ray sources: SHAP analysis of multi-wavelength features

Abstract

Extensive astronomical surveys, like those conducted with the {\em Chandra} X-ray Observatory, detect hundreds of thousands of unidentified cosmic sources. Machine learning (ML) methods offer an efficient, probabilistic approach to classify them, which can be useful for making discoveries and conducting deeper studies. In earlier work, we applied the LightGBM (ML model) to classify 277,069 {\em Chandra} point sources into eight categories: active galactic nuclei (AGN), X-ray emitting stars, young stellar objects (YSO), high-mass X-ray binaries, low-mass X-ray binaries, ultraluminous X-ray sources, cataclysmic variables, and pulsars. In this work, we present the classification table of 54,770 robustly classified sources (over $3\sigma$ confidence), including 14,066 sources at $>4\sigma$ significance. To ensure classification reliability and gain a deeper insight, we investigate the multiwavelength feature relationships learned by the LightGBM model, focusing on AGNs, Stars, and YSOs. We employ Explainable Artificial Intelligence (XAI) techniques, specifically, SHapley Additive exPlanations (SHAP), to quantify the contribution of individual features and their interactions to the predicted classification probabilities. Among other things, we find infrared-optical and X-ray decision boundaries for separating AGN/Stars, and infrared-X-ray boundaries for YSOs. These results are crucial for estimating object classes even with limited multiwavelength data. This study represents one of the earliest applications of XAI to large-scale astronomical datasets, demonstrating ML models' potential for uncovering physically meaningful patterns in data in addition to classification. Finally, our publicly available, extensive, and interactive catalogue will be helpful to explore the contributions of features and their combinations in greater detail in the future.

15 pages, 9 figures, 2 tables

References in corpus (30)

• SciPy 1.0--Fundamental Algorithms for Scientific Computing in Python
• Astropy: A Community Python Package for Astronomy
• Gaia Data Release 2. Summary of the contents and survey properties
• The Astropy Project: Building an inclusive, open-science project and status of the v2.0 core package
• The Astropy Project: Sustaining and Growing a Community-oriented Open-source Project and the Latest Major Release (v5.0) of the Core Package
• The Host Galaxies of AGN
• Interpretable Explanations of Black Boxes by Meaningful Perturbation
• Mid-Infrared Selection of AGN with the Wide-Field Infrared Survey Explorer. I. Characterizing WISE-Selected AGN in COSMOS
• A new photometric technique for the joint selection of star-forming and passive galaxies at 1.4<z<2.5
• The Chandra Deep Field-South Survey: 4 Ms Source Catalogs
• Mid-Infrared Selection of Active Galactic Nuclei with the Wide-Field Infrared Survey Explorer. II. Properties of WISE-Selected Active Galactic Nuclei in the NDWFS Boötes Field
• Spectral properties of quasars from Sloan Digital Sky Survey data release 14: The catalog
• Star-galaxy Classification Using Deep Convolutional Neural Networks
• Identification of 1.4 Million AGNs in the Mid-Infrared using WISE Data
• Chandra Deep X-ray Observation of a Typical Galactic Plane Region and Near-Infrared Identification
• Chandra Localizations and Spectra of INTEGRAL Sources in the Galactic Plane: The Cycle 9 Sample
• Quaia, the Gaia-unWISE Quasar Catalog: An All-Sky Spectroscopic Quasar Sample
• Transient-optimised real-bogus classification with Bayesian Convolutional Neural Networks -- sifting the GOTO candidate stream
• Classifying Unidentified X-ray Sources in the Chandra Source Catalog Using a Multiwavelength Machine-learning Approach
• Using Machine Learning to Link Black Hole Accretion Flows with Spatially Resolved Polarimetric Observables
• Multidimensional Data Driven Classification of Emission-line Galaxies
• Identifying preflare spectral features using explainable artificial intelligence
• Optical and infrared properties of active galactic nuclei in the Lockman Hole
• X-ray selected Infrared Excess AGN in the Chandra Deep Fields: a moderate fraction of Compton-thick sources
• Classification of 4XMM-DR9 Sources by Machine Learning
• Classification of X-ray Sources in the XMM-Newton Serendipitous Source Catalog: Objects of Special Interest
• Unsupervised Machine Learning for the Classification of Astrophysical X-ray Sources
• Deep learning interpretability analysis for carbon star identification in Gaia DR3
• Automated classification of Chandra X-ray point sources using machine learning methods
• Machine learning applications for the study of AGN physical properties using photometric observations