Time series anomaly detection with reconstruction-based state-space models
This work addresses anomaly detection for real-time monitoring in various domains, but it is incremental as it builds on existing reconstruction-based and state-space methods.
The authors tackled the problem of unsupervised anomaly detection in multivariate time series by proposing a framework that jointly learns observation and dynamic models with LSTM-based encoder-decoders and bidirectional state transitions, achieving superior performance in empirical studies on synthetic and real-world datasets.
Recent advances in digitization have led to the availability of multivariate time series data in various domains, enabling real-time monitoring of operations. Identifying abnormal data patterns and detecting potential failures in these scenarios are important yet rather challenging. In this work, we propose a novel unsupervised anomaly detection method for time series data. The proposed framework jointly learns the observation model and the dynamic model, and model uncertainty is estimated from normal samples. Specifically, a long short-term memory (LSTM)-based encoder-decoder is adopted to represent the mapping between the observation space and the latent space. Bidirectional transitions of states are simultaneously modeled by leveraging backward and forward temporal information. Regularization of the latent space places constraints on the states of normal samples, and Mahalanobis distance is used to evaluate the abnormality level. Empirical studies on synthetic and real-world datasets demonstrate the superior performance of the proposed method in anomaly detection tasks.