[论文解读] Explaining Anomalies Detected by Autoencoders Using SHAP

Liat Antwarg, Miller, Ronnie Mindlin+2 more|arXiv (Cornell University)|Mar 6, 2019

Anomaly Detection Techniques and Applications参考文献 36被引用 83

一句话总结

该论文提出一种与模型无关的 Kernel SHAP 方法来解释自编码器检测到的异常，将高重构误差与最具影响力的特征联系起来，并通过真实世界的用户研究和合成数据进行验证。

ABSTRACT

Anomaly detection algorithms are often thought to be limited because they don't facilitate the process of validating results performed by domain experts. In Contrast, deep learning algorithms for anomaly detection, such as autoencoders, point out the outliers, saving experts the time-consuming task of examining normal cases in order to find anomalies. Most outlier detection algorithms output a score for each instance in the database. The top-k most intense outliers are returned to the user for further inspection; however the manual validation of results becomes challenging without additional clues. An explanation of why an instance is anomalous enables the experts to focus their investigation on most important anomalies and may increase their trust in the algorithm. Recently, a game theory-based framework known as SHapley Additive exPlanations (SHAP) has been shown to be effective in explaining various supervised learning models. In this research, we extend SHAP to explain anomalies detected by an autoencoder, an unsupervised model. The proposed method extracts and visually depicts both the features that most contributed to the anomaly and those that offset it. A preliminary experimental study using real world data demonstrates the usefulness of the proposed method in assisting the domain experts to understand the anomaly and filtering out the uninteresting anomalies, aiming at minimizing the false positive rate of detected anomalies.

研究动机与目标

• 说明按实例进行解释的必要性，以提高专家对基于自编码器的异常检测的信任。
• 开发一个黑盒解释方法，在不知晓内部自编码器架构的情况下也能工作。
• 将高重构误差与对异常分数贡献最大的特征联系起来。
• 提供将贡献特征与抵消特征分离的可视化和表格化解释。
• 通过用户研究、合成真实标签实验、鲁棒性测试和异常分数操作来评估解释效果。

提出的方法

• 计算重构误差 L(X,X')，作为逐特征误差的平方和。
• 识别具有最高逐特征重构误差的前M个特征，以聚焦解释。
• 使用 Kernel SHAP 计算每个前特征对预测其重构值 X'i 的 SHAP 值。
• 利用 SHAP 值的极性以及 X 与 X' 的比较，将 SHAP 值分成贡献（使预测偏离真实值）和抵消（使预测趋向真实值）特征。
• 将解释呈现为一个彩色表格，显示每个前特征的贡献特征（红色）和抵消特征（蓝色），SHAP 值的大小表示重要性。
• 可选地与一种替代方法进行比较，即 SHAP 通过额外层解释总重构误差，并确认前特征的一致性。

实验结果

研究问题

• RQ1 Kernel SHAP 是否能为自编码器检测到的异常提供可靠的、模型无关的解释？
• RQ2 哪些特征及其交互最能解释自编码器输出中的高重构误差？
• RQ3 在此情境下，基于 SHAP 的解释是否比像 LIME 这样的其他方法更准确地反映真正的贡献因素？
• RQ4 解释是否提高领域专家对现实数据中异常的理解和检查效率？

主要发现

• 提出的基于 SHAP 的解释揭示了自编码器检测到的异常的贡献特征和抵消特征。
• 领域专家表示可视化解释有助于聚焦检查中最重要的解释性特征。
• 在合成真实标签测试中，解释通过 SHAP 正确识别了导致异常的确切特征。
• The SHAP-based explanations were more robust than LIME in the evaluated settings.
• The explanations were effective at reducing the anomaly score when used to manipulate explanatory features in the experiments.
• Across real-world datasets, the method supported better interpretability without requiring knowledge of the autoencoder internals.