揭示驾驶场景中动作预测的解释歧义现象
Rashomon in the Streets: Explanation Ambiguity in Scene Understanding
- 用符号化图表示法构建可解释模型集
- 两类模型间解释差异显著，达38%以上
- 适合关注AI可信度的研究者阅读
可解释人工智能（XAI）对自动驾驶等安全关键应用至关重要。然而，'拉什莫恩效应'——即多个准确率相当但解释不同的模型共存——威胁着XAI的可靠性。本文首次对真实驾驶场景中的动作预测任务进行该效应的实证量化。采用定性可解释图（QXGs）作为符号化场景表示，训练了两类模型：基于梯度提升的可解释对偶模型与复杂的图神经网络（GNNs）。通过特征归因方法，测量模型内部及跨类别的解释一致性。结果发现解释分歧显著，跨模型类别平均一致率低于62%。研究表明，解释歧义是问题本身的固有属性，而非建模误差。
原文摘要 · Abstract (English)
Explainable AI (XAI) is essential for validating and trusting models in safety-critical applications like autonomous driving. However, the reliability of XAI is challenged by the Rashomon effect, where multiple, equally accurate models can offer divergent explanations for the same prediction. This paper provides the first empirical quantification of this effect for the task of action prediction in real-world driving scenes. Using Qualitative Explainable Graphs (QXGs) as a symbolic scene representation, we train Rashomon sets of two distinct model classes: interpretable, pair-based gradient boosting models and complex, graph-based Graph Neural Networks (GNNs). Using feature attribution methods, we measure the agreement of explanations both within and between these classes. Our results reveal significant explanation disagreement. Our findings suggest that explanation ambiguity is an inherent property of the problem, not just a modeling artifact.
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。