针对少类别场景，提出更真实的解释可信度评估方法。
Few-class Fidelity: Evaluating Explanations of Real-conditions CNN classifiers with Optimized Perturbations
- 生成符合数据分布的扰动，评估解释方法可信度
- 在医疗与自然图像上验证了领域与数据对解释的影响
- 适合关注模型可解释性的研究者和开发者
卷积神经网络（CNN）在众多领域广泛应用，因其高精度和自动化速度，使用户能聚焦于更高阶任务。为理解模型并避免部署中的偏见，训练后可采用可解释人工智能（XAI）技术。然而随着XAI方法增多，其评价标准不一，缺乏共识。本文提出一种基于可信度的XAI度量变体，聚焦于实际应用中常见的少类别场景。该方法生成符合数据分布、引发不确定性的扰动，以更准确衡量XAI方法的忠实性。通过与人工主导的目标定位和分割指标对比，验证了该评估框架的有效性。在医疗与自然图像应用中，该方法揭示了领域、数据整理与XAI选择之间的复杂关联，有助于新CNN模型训练的验证。
原文摘要 · Abstract (English)
The wide use of Convolutional Neural Networks (CNN) in numerous domains and real-world classification applications is justified by their high precision and automation speed, helping users concentrate on higher-expertise tasks. To better understand the models and avoid bias during deployment, eXplainable Artificial Intelligence (XAI) techniques can be used after training. But as the list of XAI solutions expand, comparisons between them diverge, and consensus over their evaluation cannot be reached. This paper proposes a variation of Fidelity-based XAI metrics, with a focus on real-conditions applications, where the number of classes is often low. The approach generates in-distribution, uncertainty-provoking perturbations, to ensure proper measurement of the XAI methods faithfulness. As demonstration of the evaluation framework usefulness, it is compared with human-centric object localization and segmentation metrics. Once applied to both medical and natural imaging applications, it highlights the intricate correlation between domain, data curation, and XAI solution choices in order to validate training of a new CNN model.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。