Regularization-based Continual Learning for Anomaly Detection in Discrete Manufacturing

The early and robust detection of anomalies occurring in discrete manufacturing processes allows operators to prevent harm, e.g. defects in production machinery or products. While current approaches for data-driven anomaly provide good results on the exact they were trained on, often lack ability flexibly adapt changes, products. Continual learning promises such flexibility, allowing an automatic adaption previously learnt knowledge new tasks. Therefore, this article discusses different continual from group of regularization strategies, which are implemented, evaluated compared based on a real industrial metal forming dataset.