发表机构
University of Nottingham; Telespazio UK for European Space Agency; Université Paris-Saclay; Université Paris Cité; CEA; CNRS; AIM; Dr. Karl Remeis-Sternwarte & ECAP, FAU Erlangen-Nürnberg; National Observatory of Athens; INAF, IASF Milano（诺丁汉大学; 欧洲航天局英国泰莱斯帕齐奥公司; 巴黎萨克雷大学; 巴黎西岱大学; 法国原子能和替代能源委员会; 法国国家科学研究中心; 天体物理学与宇宙学研究所; 埃尔朗根-纽伦堡大学卡尔·雷梅斯天文台与应用理论物理中心; 雅典国家天文台; 意大利国家天体物理研究所米兰天文物理研究所）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究针对X射线源分类任务，对比不同维度CNN在XMM-牛顿观测数据上的表现，发现3D CNN结合时空谱信息的方法可实现更稳健的分类，且能隐式处理背景，无需手动扣除。
AI 中文摘要
区分微弱延展X射线源是X射线巡天中一项基础却极具挑战性的任务。本文针对活动星系核（AGN）和星系团的模拟XMM-牛顿卫星观测数据，开展卷积神经网络（CNN）模型的基准对比研究，重点关注数据表示的影响及该数据的可解释性。我们评估了在不同数据表示上训练的1D、2D、2D+1D和3D CNN，这些数据源自AGN和星系团的模拟XMM-牛顿观测数据，包括1D（光谱）、2D（成像）和3D（事件立方体）数据。我们还应用3D-GradCAM对学习到的特征进行解释，评估模型决策的物理基础，以突破黑箱分类的局限。结果表明，更高的光谱分辨率可提升分类器性能；使用1D和2D信息能提供可靠基准，而融合两个领域信息可显著提升诊断效果。我们发现，直接应用于时空谱事件立方体的3D CNN架构表现出最稳健的性能，在交叉验证折中具有优异的一致性。可解释性分析显示，3D模型能自主学习区分星系团内介质的弥散热辐射与AGN的局域非热幂律特征，它能利用低维投影中丢失的多维相关性。尽管其相较于2D与1D组合基准的性能提升有限，但3D方法展现出隐式表征背景的能力。该方法直接操作事件立方体内的总光子计数，无需手动背景扣除与建模，为大规模星表中X射线源的自动分类提供了一致、简化的端到端流程。
英文摘要
Distinguishing between faint extended X-ray sources is a fundamental yet challenging task in X-ray surveys. This paper presents a benchmark comparison of convolutional neural network models for classifying simulated XMM-Newton observations of AGN and galaxy clusters, focusing on the impact of data representation and interpretability of that data. We evaluate 1D, 2D, 2D+1D, and 3D CNNs trained on distinct data representations derived from simulated XMM-Newton observations of AGN and clusters: 1D (spectral), 2D (imaging) and 3D (event cube) data. We further apply 3D-GradCAM to interpret the learned features and assess the physical basis for the model's decision making to move beyond black-box classification. Our results demonstrate that increased spectral resolution enhances classifier performance using 1D and 2D information provides a strong baseline, the integration of both domains yields significant diagnostic gains. We find that a 3D CNN architecture applied directly to spatio-spectral event cubes provides the most robust performance, with superior consistency across cross-validation folds. Interpretability analysis reveals that the 3D model autonomously learns to distinguish the diffuse thermal emission of the intracluster medium from the localized, non-thermal power-law signatures of AGN. It exploits multi-dimensional correlations that are lost in lower-dimensional projections. While the performance gain over combined 2D and 1D baselines is marginal, the 3D approach demonstrates an ability to perform implicit background characterisation. By operating directly on total photon counts within event cubes, this method bypasses the need for manual background subtraction and modelling, offering a consistent, streamlined, end-to-end pipeline for automated classification of X-ray sources in large-scale catalogues.
