Anomaly Detection and Interpretation using Multimodal Autoencoder and Sparse Optimization
This work addresses the need for interpretable anomaly detection in large ICT systems to help operators localize issues, though it is incremental as it builds on existing autoencoder techniques.
The paper tackled the problem of interpreting anomalies detected by autoencoders in ICT systems by proposing a sparse optimization algorithm to identify contributing dimensions and a multimodal autoencoder to learn cross-domain relationships, achieving superior performance in specifying anomaly sources and detecting anomalies compared to conventional methods.
Automated anomaly detection is essential for managing information and communications technology (ICT) systems to maintain reliable services with minimum burden on operators. For detecting varying and continually emerging anomalies as differences from normal states, learning normal relationships inherent among cross-domain data monitored from ICT systems is essential. Deep-learning-based anomaly detection using an autoencoder (AE) is therefore promising for such complicated learning; however, its interpretation is still problematic. Since the dimensions of the input data contributing to the detected anomaly are not directly indicated in an AE, they are not suitable for localizing anomalies in large ICT systems composed of a huge amount of equipment. We propose an algorithm using sparse optimization for estimating contributing dimensions to anomalies detected with AEs. We also propose a multimodal AE (MAE) for effectively learning the relationships among cross-domain data, which can induce nonlinearity and differences in learnability among data types. We evaluated our algorithms with several datasets including real measured data in comparison with conventional algorithms and confirmed the superiority of our estimation algorithm in specifying contributing dimensions of anomalous data and our MAE in detecting anomalies in cross-domain data.