现有时间序列解释方法因计算方式错误，无法区分直接与间接影响。
The Failures of Marginal Influence-Based Attribution Methods for Global Time Series Explanations
- 用边际条件或离域梯度计算影响，混淆了直接与间接依赖关系。
- 标准方法如SHAP在自相关下不满足图忠实性，结果不可靠。
- 适合研究模型可解释性、时间序列建模的学者关注。
时间序列模型的可解释性方法主要生成平面化的归因分数：以标量形式量化某特征在特定时间点的直接影响。我们证明，此类方法的主要失败原因并非标量格式本身，而是根本性的计算错配：现有方法通过边际条件或离域梯度计算得分，在自相关条件下会混淆直接时间依赖与中介依赖。我们定义了图忠实性（DAG-faithfulness）：若解释所编码的时间依赖图与模型隐式学习的时序有向无环图（DAG）马尔可夫等价，则该解释为图忠实。特别地，我们观察到标准归因方法（如SHAP）通常不满足图忠实性，近期的时间序列感知扩展也继承了相同的计算局限。
原文摘要 · Abstract (English)
Explainability methods for time series models predominantly produce flat attribution scores: they quantify the direct influence of a feature at a timestamp by a scalar. We prove that the dominant failure mode of such methods is not the scalar format itself but a fundamental computational mismatch: existing methods compute scores via marginal conditioning or off-manifold gradients, both of which conflate direct temporal dependencies with mediated ones under autocorrelation. We also define DAG-faithfulness: an explanation is DAG-faithful if the temporal dependency graph it encodes is Markov-equivalent to the temporal directed acyclic graph (DAG) implicitly learned by the model. Particularly, we observe that standard attribution methods, specifically SHAP, are not DAG-faithful in general, and that recent time-series-aware extensions inherit the same computational limitation.
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。