Anomaly Transformer: Time Series Anomaly Detection with Association Discrepancy

by   Jiehui Xu, et al.

Unsupervisedly detecting anomaly points in time series is challenging, which requires the model to learn informative representations and derive a distinguishable criterion. Prior methods mainly detect anomalies based on the recurrent network representation of each time point. However, the point-wise representation is less informative for complex temporal patterns and can be dominated by normal patterns, making rare anomalies less distinguishable. We find that in each time series, each time point can also be described by its associations with all time points, presenting as a point-wise distribution that is more expressive for temporal modeling. We further observe that due to the rarity of anomalies, it is harder for anomalies to build strong associations with the whole series and their associations shall mainly concentrate on the adjacent time points. This observation implies an inherently distinguishable criterion between normal and abnormal points, which we highlight as the Association Discrepancy. Technically we propose the Anomaly Transformer with an Anomaly-Attention mechanism to compute the association discrepancy. A minimax strategy is devised to amplify the normal-abnormal distinguishability of the association discrepancy. Anomaly Transformer achieves state-of-the-art performance on six unsupervised time series anomaly detection benchmarks for three applications: service monitoring, space & earth exploration, and water treatment.


page 1

page 2

page 3

page 4


Memory-augmented Adversarial Autoencoders for Multivariate Time-series Anomaly Detection with Deep Reconstruction and Prediction

Detecting anomalies for multivariate time-series without manual supervis...

Reconstruct Anomaly to Normal: Adversarial Learned and Latent Vector-constrained Autoencoder for Time-series Anomaly Detection

Anomaly detection in time series has been widely researched and has impo...

DCdetector: Dual Attention Contrastive Representation Learning for Time Series Anomaly Detection

Time series anomaly detection is critical for a wide range of applicatio...

TiSAT: Time Series Anomaly Transformer

While anomaly detection in time series has been an active area of resear...

AnomalyBERT: Self-Supervised Transformer for Time Series Anomaly Detection using Data Degradation Scheme

Mechanical defects in real situations affect observation values and caus...

Focus the Discrepancy: Intra- and Inter-Correlation Learning for Image Anomaly Detection

Humans recognize anomalies through two aspects: larger patch-wise repres...

High-performance Time Series Anomaly Discovery on Graphics Processors

Currently, discovering subsequence anomalies in time series remains one ...

Please sign up or login with your details

Forgot password? Click here to reset