Revisiting VAE for Unsupervised Time Series Anomaly Detection: A Frequency Perspective
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866929234290671616 |
|---|---|
| author | Wang, Zexin Pei, Changhua Ma, Minghua Wang, Xin Li, Zhihan Pei, Dan Rajmohan, Saravan Zhang, Dongmei Lin, Qingwei Zhang, Haiming Li, Jianhui Xie, Gaogang |
| author_facet | Wang, Zexin Pei, Changhua Ma, Minghua Wang, Xin Li, Zhihan Pei, Dan Rajmohan, Saravan Zhang, Dongmei Lin, Qingwei Zhang, Haiming Li, Jianhui Xie, Gaogang |
| contents | Time series Anomaly Detection (AD) plays a crucial role for web systems. Various web systems rely on time series data to monitor and identify anomalies in real time, as well as to initiate diagnosis and remediation procedures. Variational Autoencoders (VAEs) have gained popularity in recent decades due to their superior de-noising capabilities, which are useful for anomaly detection. However, our study reveals that VAE-based methods face challenges in capturing long-periodic heterogeneous patterns and detailed short-periodic trends simultaneously. To address these challenges, we propose Frequency-enhanced Conditional Variational Autoencoder (FCVAE), a novel unsupervised AD method for univariate time series. To ensure an accurate AD, FCVAE exploits an innovative approach to concurrently integrate both the global and local frequency features into the condition of Conditional Variational Autoencoder (CVAE) to significantly increase the accuracy of reconstructing the normal data. Together with a carefully designed "target attention" mechanism, our approach allows the model to pick the most useful information from the frequency domain for better short-periodic trend construction. Our FCVAE has been evaluated on public datasets and a large-scale cloud system, and the results demonstrate that it outperforms state-of-the-art methods. This confirms the practical applicability of our approach in addressing the limitations of current VAE-based anomaly detection models. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_02820 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Revisiting VAE for Unsupervised Time Series Anomaly Detection: A Frequency Perspective Wang, Zexin Pei, Changhua Ma, Minghua Wang, Xin Li, Zhihan Pei, Dan Rajmohan, Saravan Zhang, Dongmei Lin, Qingwei Zhang, Haiming Li, Jianhui Xie, Gaogang Machine Learning Time series Anomaly Detection (AD) plays a crucial role for web systems. Various web systems rely on time series data to monitor and identify anomalies in real time, as well as to initiate diagnosis and remediation procedures. Variational Autoencoders (VAEs) have gained popularity in recent decades due to their superior de-noising capabilities, which are useful for anomaly detection. However, our study reveals that VAE-based methods face challenges in capturing long-periodic heterogeneous patterns and detailed short-periodic trends simultaneously. To address these challenges, we propose Frequency-enhanced Conditional Variational Autoencoder (FCVAE), a novel unsupervised AD method for univariate time series. To ensure an accurate AD, FCVAE exploits an innovative approach to concurrently integrate both the global and local frequency features into the condition of Conditional Variational Autoencoder (CVAE) to significantly increase the accuracy of reconstructing the normal data. Together with a carefully designed "target attention" mechanism, our approach allows the model to pick the most useful information from the frequency domain for better short-periodic trend construction. Our FCVAE has been evaluated on public datasets and a large-scale cloud system, and the results demonstrate that it outperforms state-of-the-art methods. This confirms the practical applicability of our approach in addressing the limitations of current VAE-based anomaly detection models. |
| title | Revisiting VAE for Unsupervised Time Series Anomaly Detection: A Frequency Perspective |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2402.02820 |