发表机构
National Higher School of Computer Science (ESI); Université Claude Bernard Lyon 1（国家高等计算机科学学院（ESI）; 里昂第一大学（克洛德·贝尔纳里昂第一大学））
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究针对AddGraph框架提出双时空归因机制的事后可解释性框架X-AddGraph，在不损失检测性能的前提下为动态图边级异常检测提供可解释性，其长期归因能力优于空间盲解释器。
AI 中文摘要
针对动态图异常检测的深度学习检测器已达到较高准确率，但仍存在不透明性问题：当某条边被标记为异常时，分析人员仅能得到一个分数，却无法获知原因。这种不透明性在部署此类检测器的协作式、受监管信息系统中是不可接受的，此类系统要求自动化决策具备可审计性与可信度。我们针对AddGraph解决这一缺口，AddGraph是用于动态图边级异常检测的基础GCN+GRU框架，据我们所知，该框架从未配备任何形式的可解释性。我们提出了一种严格的事后可解释性框架X-AddGraph，其构建于双时空归因（DSTA）机制之上，该机制的三个组件分别与AddGraph的架构模块对齐：针对当前邻接结构的基于梯度的相关性归因（空间维度）、推理过程中已计算的上下文注意力权重的直接读取（短期时间维度，零额外成本）、通过循环隐藏状态进行的梯度回滚（长期时间维度）。由于检测器保持冻结状态，检测性能完全保留（AUC变化量为0，经实证验证精确至小数点后十位）。在UCI Message基准测试中，我们训练的AddGraph基线达到平均每快照AUC为0.8705，超过了原发表结果；X-AddGraph在完全复现所有分数的同时，为原本无解释的情况添加了解释。针对四类边群体——高置信度真阳性、低置信度真阳性、假阳性和随机样本——进行评估时，长期归因识别出的携带反事实信号的历史快照数量显著多于随机选择（0.127 vs. 0.074），这是任何空间盲解释器都无法提供的能力。我们发布了实现代码以支持完全可复现。
英文摘要
Deep learning detectors for anomalies in dynamic graphs have reached strong accuracy, yet they remain opaque: when an edge is flagged, the analyst receives a score but no reason. This opacity is untenable in the cooperative, regulated information systems where such detectors are deployed, where automated decisions must be auditable and trustworthy. We address this gap for AddGraph, the foundational GCN+GRU framework for edge-level anomaly detection in dynamic graphs, which to our knowledge has never been equipped with any form of explainability. We present a strictly post-hoc explainability framework, X-AddGraph, built on a Dual Spatial-Temporal Attribution (DSTA) mechanism whose three components are each aligned with one of AddGraph's architectural modules: a gradient-based relevance attribution over the current adjacency structure (spatial), a direct reading of the contextual attention weights already computed during inference (short-term temporal, at zero additional cost), and a gradient rollback through the recurrent hidden states (long-term temporal). Because the detector is frozen, detection performance is preserved exactly (Delta AUC = 0, verified empirically to ten decimal places). On the UCI Message benchmark, our trained AddGraph baseline reaches an average per-snapshot AUC of 0.8705, exceeding the originally published result; X-AddGraph reproduces every score identically while adding explanations where none existed. Evaluated across four edge populations - confident true positives, low-confidence true positives, false positives, and random samples - the long-term attribution identifies historical snapshots carrying significantly more counterfactual signal than random selection (0.127 vs. 0.074), a capability that no spatially-blind explainer can provide. We release our implementation for full reproducibility.
CommentsUnder Review : The International Conference on Cooperative Information Systems (CoopIS)
