An Evaluation of Anomaly Detection and Diagnosis in Multivariate Time Series

Abstract

Several techniques for multivariate time series anomaly detection have been proposed recently, but a systematic comparison on a common set of datasets and metrics is lacking. This paper presents a systematic and comprehensive evaluation of unsupervised and semi-supervised deep-learning based methods for anomaly detection and diagnosis on multivariate time series data from cyberphysical systems. Unlike previous works, we vary the model and post-processing of model errors, i.e. the scoring functions independently of each other, through a grid of 10 models and 4 scoring functions, comparing these variants to state of the art methods. In time-series anomaly detection, detecting anomalous events is more important than detecting individual anomalous time-points. Through experiments, we find that the existing evaluation metrics either do not take events into account, or cannot distinguish between a good detector and trivial detectors, such as a random or an all-positive detector. We propose a new metric to overcome these drawbacks, namely, the composite F-score ($F{c}_1$), for evaluating time-series anomaly detection. Our study highlights that dynamic scoring functions work much better than static ones for multivariate time series anomaly detection, and the choice of scoring functions often matters more than the choice of the underlying model. We also find that a simple, channel-wise model - the Univariate Fully-Connected Auto-Encoder, with the dynamic Gaussian scoring function emerges as a winning candidate for both anomaly detection and diagnosis, beating state of the art algorithms.

arXiv commentIEEE Transactions on Neural Networks and Learning Systems

References in corpus (12)

• An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling
• Deep Learning for Anomaly Detection: A Review
• Detecting Spacecraft Anomalies Using LSTMs and Nonparametric Dynamic Thresholding
• Tutorial on Variational Autoencoders
• Deep Learning for Anomaly Detection: A Survey
• Unsupervised Anomaly Detection via Variational Auto-Encoder for Seasonal KPIs in Web Applications
• Time-Series Anomaly Detection Service at Microsoft
• LSTM-based Encoder-Decoder for Multi-sensor Anomaly Detection
• Current Time Series Anomaly Detection Benchmarks are Flawed and are Creating the Illusion of Progress
• MAD-GAN: Multivariate Anomaly Detection for Time Series Data with Generative Adversarial Networks
• Automatic Anomaly Detection in the Cloud Via Statistical Learning
• A Joint Model for IT Operation Series Prediction and Anomaly Detection

Cited by in corpus (10)

• Deep Learning for Time Series Anomaly Detection: A Survey
• TFAD: A Decomposition Time Series Anomaly Detection Architecture with Time-Frequency Analysis
• Correlation-aware Spatial-Temporal Graph Learning for Multivariate Time-series Anomaly Detection
• Local Evaluation of Time Series Anomaly Detection Algorithms
• CARE to Compare: A real-world dataset for anomaly detection in wind turbine data
• Angel or Devil: Discriminating Hard Samples and Anomaly Contaminations for Unsupervised Time Series Anomaly Detection
• Entropy Causal Graphs for Multivariate Time Series Anomaly Detection
• Time series anomaly detection with reconstruction-based state-space models
• SoK: Evaluations in Industrial Intrusion Detection Research
• TriP-LLM: A Tri-Branch Patch-wise Large Language Model Framework for Time-Series Anomaly Detection