.png){kind=icon}
.png){kind=icon}
Return
Situation-Aware Multivariate Time Series Anomaly Detection Through Active Learning and Contrast VAE-Based Models in Large Distributed Systems
.png){kind=icon}
PRE
AI
DOI：10.1109/JSAC.2022.3191341.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The massive amounts of monitoring data in network applications bring an urgent need for intelligent operation in large distributed systems. The key problem is precisely detecting anomalies in multivariate time series (MTS) monitoring metrics with the awareness of different application scenarios. Unsupervised NITS anomaly detection methods aim at detecting data anomalies from historical MTS without considering the out-of-band information (including user feedback and background information like code deployment status), which leads to poor performance in practice. To take advantage of the out-of-band information, we propose ACVAE, an MTS anomaly detection algorithm through active learning and contrast VAE-based detection models, which simultaneously learns NITS data's normal and anomalous patterns for anomaly detection. We also use a learnable prior to capture system status from the background information. Moreover, we propose a query model for VAE-based methods, which can learn to query labels of the must useful instances to train the detection model. We evaluate our algorithm on three different monitoring situations in eBay's search back-end systems. ACVAE achieves a range F1 score of 0.68 similar to 0.96 with only 3% labels, significantly outperforming the best competing methods by 0.18 similar to 0.50, and even better than a supervised ensemble method designed by domain experts in eBay.
Keywords:
Anomaly detection
out-of-band information
multivariate time series
active learning
variational autoencoder
Journal
IF:
17.2
Papers:
6.4K
Citations:
3.1W