Recent semi-supervised anomaly detection methods that are trained using small labeled anomaly examples and large unlabeled data (mostly normal data) have shown largely improved performance over unsupervised methods. However, these methods often focus on fitting abnormalities illustrated by the given anomaly examples only (i.e.,, seen anomalies), and consequently they fail to generalize to those that are not, i.e., new types/classes of anomaly unseen during training. To detect both seen and unseen anomalies, we introduce a novel deep weakly-supervised approach, namely Pairwise Relation prediction Network (PReNet), that learns pairwise relation features and anomaly scores by predicting the relation of any two randomly sampled training instances, in which the pairwise relation can be anomaly-anomaly, anomaly-unlabeled, or unlabeled-unlabeled. Since unlabeled instances are mostly normal, the relation prediction enforces a joint learning of anomaly-anomaly, anomaly-normal, and normal-normal pairwise discriminative patterns, respectively. PReNet can then detect any seen/unseen abnormalities that fit the learned pairwise abnormal patterns, or deviate from the normal patterns. Further, this pairwise approach also seamlessly and significantly augments the training anomaly data. Empirical results on 12 real-world datasets show that PReNet significantly outperforms nine competing methods in detecting seen and unseen anomalies. We also theoretically and empirically justify the robustness of our model w.r.t. anomaly contamination in the unlabeled data. The code is available at https://github.com/mala-lab/PReNet.

翻译：近期，利用少量标注异常样本和大量未标注数据（主要为正常数据）进行训练的半监督异常检测方法，其性能已显著超越无监督方法。然而，这些方法通常仅专注于拟合给定异常样本所呈现的异常模式（即"已见异常"），因而难以泛化至未出现的异常类型（即训练过程中未见的新类别异常）。为同时检测已见与未见异常，我们提出一种新颖的深度弱监督方法——成对关系预测网络（PReNet），该方法通过预测任意两个随机采样训练实例之间的成对关系，学习成对关系特征与异常分数。其中，成对关系可分为异常-异常、异常-未标注或未标注-未标注三类。由于未标注实例大多为正常样本，这种关系预测分别强制实现了异常-异常、异常-正常与正常-正常成对判别模式的联合学习。PReNet进而能够检测任何符合所学习异常模式或偏离正常模式的已见/未见异常。此外，这种成对方法还能无缝且显著地扩充训练异常数据。在12个真实数据集上的实验结果表明，PReNet在检测已见与未见异常方面显著优于九种对比方法。我们还从理论上和实证上证明了模型对未标注数据中异常污染的鲁棒性。代码已开源至https://github.com/mala-lab/PReNet。