AI 中文总结
本研究针对多标签分类任务，在MS-COCO数据集上训练VGG16、ResNet50模型，应用CRP、CRAFT方法生成概念级解释，揭示CXAI可凸显DNNs弱点、降低混淆、暴露数据集偏差，助力模型泛化理解与偏差诊断。
AI 中文摘要
部署在医疗保健和自动驾驶等高风险领域的深度神经网络（DNNs）不仅需要准确，还需具备可解释性以确保用户信任。在真实世界的计算机视觉任务中，这些模型通常处理包含背景噪声且标注密集的复杂图像。为使此类模型具备可解释性，需评估基于概念的可解释人工智能（CXAI）方法的适用性和问题解决能力。本研究在多标签分类场景下探索CXAI的应用案例，在MS-COCO数据集（Microsoft Common Objects in Context）的20个标注最密集的标签上训练VGG16和ResNet50两个DNNs；应用CRP（Concept Relevance Propagation，概念相关性传播）和CRAFT（Concept Recursive Activation FacTorization，概念递归激活因子分解）两种CXAI方法生成概念级解释，并开展整体评估。分析揭示三项关键发现：（1）CXAI可凸显DNNs的学习弱点；（2）更高的概念独特性会降低标签与概念的混淆；（3）环境概念会暴露数据集引发的偏差。结果表明，CXAI有潜力增强对模型泛化能力的理解，并诊断由数据集引发的偏差。
英文摘要
Deep Neural Networks (DNNs) deployed in high-risk domains, such as healthcare and autonomous driving, must be not only accurate but also understandable to ensure user trust. In real-world computer vision tasks, these models often operate on complex images containing background noise and are heavily annotated. To make such models explainable, Concept-based Explainable AI (CXAI) methods need to be assessed for their applicability and problem-solving capacity. In this work, we explore CXAI use cases in multi-label classification by training two DNNs, VGG16 and ResNet50, on the 20 most annotated labels in the MS-COCO dataset (Microsoft Common Objects in Context). We apply two CXAI methods, CRP (Concept Relevance Propagation) and CRAFT (Concept Recursive Activation FacTorization), to generate concept-level explanations and investigate the overall evaluations. Our analysis reveals three key findings: (1) CXAI highlights learning weaknesses in DNNs, (2) higher concept distinctiveness reduces label and concept confusion, and (3) environmental concepts expose dataset-induced biases. Our results demonstrate the potential of CXAI to enhance the understanding of model generalizability and to diagnose bias instigated by the dataset.
CommentsPublished at EXPLAINABILITY2025
