自解释图神经网络的解释可能完全无关，导致误判风险。
GNN Explanations that do not Explain and How to find Them
- 发现自解释GNN的解释可与真实推理过程无关
- 多数忠实度指标无法识别此类失效情况
- 提出新度量可可靠检测恶意或自然出现的虚假解释
自解释图神经网络（SE-GNNs）的解释对理解模型内部机制和识别敏感属性滥用至关重要。尽管已有研究指出其解释可能次优甚至具有误导性，但缺乏对失败模式的系统刻画。本文揭示了SE-GNN解释的关键缺陷：解释可能与模型预测标签的真正推理过程完全无关。我们证明，许多SE-GNN可在实现最优真实风险的同时生成这种退化解释；同时，现有大多数忠实度度量无法识别此类失效模式。实证分析显示，退化解释既可被恶意植入（如隐藏敏感属性使用），也可自然产生，凸显了可靠审计的必要性。为此，我们提出一种新型忠实度度量，在恶意与自然场景下均能可靠标记退化解释为不忠实。代码已开源：https://github.com/steveazzolin/gnn_deg_expl。
原文摘要 · Abstract (English)
Explanations provided by Self-explainable Graph Neural Networks (SE-GNNs) are fundamental for understanding the model's inner workings and for identifying potential misuse of sensitive attributes. Although recent works have highlighted that these explanations can be suboptimal and potentially misleading, a characterization of their failure cases is unavailable. In this work, we identify a critical failure of SE-GNN explanations: explanations can be unambiguously unrelated to how the SE-GNNs infer labels. We show that, on the one hand, many SE-GNNs can achieve optimal true risk while producing these degenerate explanations, and on the other, most faithfulness metrics can fail to identify these failure modes. Our empirical analysis reveals that degenerate explanations can be maliciously planted (allowing an attacker to hide the use of sensitive attributes) and can also emerge naturally, highlighting the need for reliable auditing. To address this, we introduce a novel faithfulness metric that reliably marks degenerate explanations as unfaithful, in both malicious and natural settings. Our code is available at https://github.com/steveazzolin/gnn_deg_expl.
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。