Chen Qiu, Aodong Li, Marius Kloft, Maja Rudolph, Stephan Mandt
TL;DR本研究提出一种在未标记异常情况下训练异常检测器的策略,通过联合推断二进制标签(正常 vs. 异常)并更新模型参数来使用两个损失的组合,表现出比基准测试更显著的改进。
Abstract
anomaly detection aims at identifying data points that show systematic
deviations from the majority of data in an unlabeled dataset. A common
assumption is that clean training data (free of anomalies) is available, which
is often violated in practice. We propose a strategy for training