Robust Anomaly Detection in Feature-Evolving Time Series. (5th January 2021)

• Record Type: Journal Article
• Authors: Wambura, Stephen; Huang, Jianbin; Li, He
• Is Part Of: Computer journal. Volume 65:Number 5(2022)
• Page Start: 1242
• Page End: 1256
• Publication Date: 2021-01-05
• Subjects: anomaly detection -- time series -- neural networks -- deep learning; Computers -- Periodicals; 005.1
• DOI: 10.1093/comjnl/bxaa174
• Languages: English
• ISSNs: 0010-4620

Abstract

Abstract: This paper addresses the anomaly detection problem in feature-evolving systems such as server machines, cyber security, financial markets and so forth where in every millisecond, $N$ -dimensional feature-evolving heterogeneous time series are generated. However, due to stochasticity and uncertainty in evolving heterogeneous time series coupled with temporal dependencies, their anomaly detection are extremely challenging. Furthermore, it is practically impossible to train an anomaly detection model per single time series across millions of metrics, leave alone memory space required to maintain the model and evolving data points in memory for timely processing in feature-evolving data streams. Thus, this paper proposes $\underline{o}$ ne sketch $\underline{f}$ its all $\underline{a}$ lgorithm (OFA), which is a real-time stochastic recurrent deep neural network anomaly detector built on assumption-free probabilistic conditional quantile regression with well-calibrated predictive uncertainty estimates. The proposed framework is capable of detecting anomalies robustly, accurately and efficiently in real time while handling randomness and variabilities in feature-evolving heterogeneous time series. Extensive experiments and rigorous evaluation on large-scale real-world data sets showcase that OFA outperforms other competitive state-of-the-art anomaly detector methods.