Unsupervised deep learning techniques are widely used to identify anomalous behaviour.The performance of such methods is a product of the amount of training data and the model size.However, the size is often a limiting factor for the deployment on resource-constrained devices.We present a novel procedure based on knowledge distillation for compressing an unsupervised anomaly detection model into a supervised deployable one and we suggest a set of techniques to improve the detection sensitivity.Compressed models perform comparably to their larger counterparts while significantly reducing the size and memory footprint.

Exploiting the rapid advances in probabilistic inference, in particular variational Bayes and variational autoencoders (VAEs), for anomaly detection (AD) tasks remains an open research question. Previous works argued that training VAE models only with inliers is insufficient and the framework should be significantly modified in order to discriminate the anomalous instances. In this work, we exploit the deep conditional variational autoencoder (CVAE) and we define an original loss function together with a metric that targets hierarchically structured data AD. Our motivating application is a real world problem: monitoring the trigger system which is a basic component of many particle physics experiments at the CERN Large Hadron Collider (LHC). In the experiments we show the superior performance of this method for classical machine learning (ML) benchmarks and for our application.