TranAD: Deep Transformer Networks for Anomaly Detection in Multivariate Time Series Data

By Shreshth Tuli, Giuliano Casale, Nicholas R. Jennings. Proceedings of the VLDB Endowment, 2022

1,012 citations recorded by the source corpus.

Efficient anomaly detection and diagnosis in multivariate time-series data is of great importance for modern industrial applications. However, building a system that is able to quickly and accurately pinpoint anomalous observations is a challenging problem. This is due to the lack of anomaly labels, high data volatility and the demands of ultra-low inference times in modern applications. Despite the recent developments of deep learning approaches for anomaly detection, only a few of them can address all of these challenges. In this paper, we propose TranAD, a deep transformer network based anomaly detection and diagnosis model which uses attention-based sequence encoders to swiftly perform inference with the knowledge of the broader temporal trends in the data. TranAD uses focus score-based self-conditioning to enable robust multi-modal feature extraction and adversarial training to gain stability. Additionally, model-agnostic meta learning (MAML) allows us to train the model using limited data. Extensive empirical studies on six publicly available datasets demonstrate that TranAD can outperform state-of-the-art baseline methods in detection and diagnosis performance with data and time-efficient training. Specifically, TranAD increases F1 scores by up to 17%, reducing training times by up to 99% compared to the baselines.

Subfields: Artificial Intelligence, Computer Networks and Communications, Signal Processing

View Paper: https://www.semanticscholar.org/paper/2d57a3f90adf3fc28f0de61fb4b7b34bccb1b92d

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Junchi Yan
2. Ziqing Ma
3. Chao Zhang
4. Yongjun Wang
5. Weiqiu Chen
6. Liang Sun
7. Yijie Wang
8. Hongzuo Xu

In Computer Networks and Communications

1. Yongjun Wang
2. Liang Sun
3. Yijie Wang
4. Hongzuo Xu
5. Guansong Pang
6. Qingsong Wen
7. Tian Zhou

In Signal Processing

1. Junchi Yan
2. Ziqing Ma
3. Yongjun Wang
4. Liang Sun
5. Yijie Wang
6. Hongzuo Xu
7. Guansong Pang
8. Qingsong Wen

Citations in the corpus

• Deep Isolation Forest for Anomaly Detection (2022, methodology): “GLocalKD [40] is a graph-level anomaly detector, and TranAD [41] is a deep anomaly detection method specifically designed for time series.”
• Deep Learning for Time Series Anomaly Detection: A Survey (2022, neutral): “However, TranAD’s adversarial training amplifies reconstruction errors to fix this.”
• Transformers in Time Series: A Survey (2022, methodology): “Besides, in multiple studies, including TranAD [Tuli et al. , 2022], MT-RVAE [Wang et al. , 2022], and TransAnomaly [Zhang et al. , 2021], researchers proposed to combine Transformer with neural generative models, such as VAEs [Kingma and Welling, 2014] and GANs [Good-fellow et al. , 2014], for…”