GDformer: Going Beyond Subsequence Isolation for Multivariate Time Series Anomaly Detection
This work addresses the problem of detecting anomalies in multivariate time series for applications like monitoring and security, offering a novel approach that improves over existing methods but is incremental in advancing the field.
The paper tackles the challenge of unsupervised anomaly detection in multivariate time series by proposing GDformer, which uses a global dictionary-enhanced transformer to derive a unified series-level detection criterion, achieving state-of-the-art performance on five benchmark datasets.
Unsupervised anomaly detection of multivariate time series is a challenging task, given the requirements of deriving a compact detection criterion without accessing the anomaly points. The existing methods are mainly based on reconstruction error or association divergence, which are both confined to isolated subsequences with limited horizons, hardly promising unified series-level criterion. In this paper, we propose the Global Dictionary-enhanced Transformer (GDformer) with a renovated dictionary-based cross attention mechanism to cultivate the global representations shared by all normal points in the entire series. Accordingly, the cross-attention maps reflect the correlation weights between the point and global representations, which naturally leads to the representation-wise similarity-based detection criterion. To foster more compact detection boundary, prototypes are introduced to capture the distribution of normal point-global correlation weights. GDformer consistently achieves state-of-the-art unsupervised anomaly detection performance on five real-world benchmark datasets. Further experiments validate the global dictionary has great transferability among various datasets. The code is available at https://github.com/yuppielqx/GDformer.