在工业质检、安全监控等领域，多类别无监督异常检测（Multi-class Unsupervised Anomaly Detection, MUAD）一直是个极具挑战的课题。传统的做法通常是训练一个复杂的编码器-解码器模型，试图重建正常样本的特征。但你有没有想过，这种费时费力的“训练”过程，真的是必须的吗？ 最近，来自湖南大学和英国格拉斯哥大学的研究团队在一篇论文中发出了灵魂拷问：Is Traini ...

Dr. James McCaffrey from Microsoft Research presents a complete program that uses the Python language LightGBM system to create a custom autoencoder for data anomaly detection. You can easily adapt ...

Anomaly detection can be powerful in spotting cyber incidents, but experts say CISOs should balance traditional signature-based detection with more bespoke methods that can identify malicious activity ...