Practical Approach to Asynchronous Multivariate Time Series Anomaly Detection and Localization

By Ahmed Abdulaal, Zhuanghua Liu, Tomer Lancewicki. Knowledge Discovery and Data Mining, 2021

498 citations recorded by the source corpus.

Engineers at eBay utilize robust methods in monitoring IT system signals for anomalies. However, the growing scale of signals, both in volumes and dimensions, overpowers traditional statistical state-space or supervised learning tools. Thus, state-of-the-art methods based on unsupervised deep learning are sought in recent research. However, we experienced flaws when implementing those methods, such as requiring partial supervision and weaknesses to high dimensional datasets, among other reasons discussed in this paper. We propose a practical approach for inferring anomalies from large multivariate sets. We observe an abundance of time series in real-world applications, which exhibit asynchronous and consistent repetitive variations, such as IT, weather, utility, and transportation. Our solution is designed to leverage this behavior. The solution utilizes spectral analysis on the latent representation of a pre-trained autoencoder to extract dominant frequencies across the signals, which are then used in a subsequent network that learns the phase shifts across the signals and produces a synchronized representation of the raw multivariate. Random subsets of the synchronous multivariate are then fed into an array of autoencoders learning to minimize the quantile reconstruction losses, which are then used to infer and localize anomalies based on a majority vote. We benchmark this method against state-of-the-art approaches on public datasets and eBay's data using their referenced evaluation methods. Furthermore, we address the limitations of the referenced evaluation methods and propose a more realistic evaluation method.

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

View Paper: https://www.semanticscholar.org/paper/5760e1ea941cf955a3f26a204ba00c0d7299e389

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. Mingsheng Long
2. Jianmin Wang
3. Donghao Luo
4. Xue Wang
5. Haixu Wu
6. Yiyuan Yang
7. Liang Sun
8. Jiehui Xu

In Computer Networks and Communications

1. Haixu Wu
2. Liang Sun
3. Jiehui Xu
4. Jianmin Wang
5. Qiang Xu
6. Zhijian Xu
7. Donghao Luo
8. Xue Wang

In Signal Processing

1. Qiang Xu
2. Ailing Zeng
3. Donghao Luo
4. Xue Wang
5. Haixu Wu
6. Tian Zhou
7. Liang Sun
8. Zhijian Xu

Citations in the corpus

• DCdetector: Dual Attention Contrastive Representation Learning for Time Series Anomaly Detection (2023, methodology): “(a) Window size [1] [3] [5] [1,3] [1,5] [3,5] [1,3,5] Multi-scale Patch-sizes 91 92 93 94 95 96 97 98”
• ModernTCN: A Modern Pure Convolution Structure for General Time Series Analysis (2024, methodology): “For anomaly detection, we compare models on five widely-used benchmarks: SMD (Su et al., 2019), SWaT (Mathur & Tippenhauer, 2016), PSM (Abdulaal et al., 2021), MSL and SMAP (Hundman et al., 2018).”
• Anomaly Transformer: Time Series Anomaly Detection with Association Discrepancy (2021, methodology): “(2) PSM (Pooled Server Metrics, Abdulaal et al. (2021)) is collected internally from multiple application server nodes at eBay with 26 dimensions. (3) Both MSL (Mars Science Laboratory rover) and SMAP (Soil Moisture Active Passive satellite) are public datasets from NASA (Su et al., 2019b) with 55 a”
• TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysis (2022, methodology): “, 2018), SWaT (Mathur & Tippenhauer, 2016), PSM (Abdulaal et al., 2021), covering service monitoring, space & earth exploration, and water treatment applications.”
• One Fits All: Power General Time Series Analysis by Pretrained LM (2023, methodology): “XGBoost Chen & Guestrin (2016), Rocket Dempster et al. (2020), LSTNet Lai et al. (2018), LSSL Gu et al. (2021), Pyraformer Liu et al. (2021), TCN Franceschi et al. (2019) and Flowformer Huang et al. (2022) are used for classification.”
• Deep Learning for Time Series Anomaly Detection: A Survey (2022, unknown): “RANSysCoders [4] outlines a real-time anomaly detection method used by eBay.”
• FITS: Modeling Time Series with 10k Parameters (2023, methodology): “We use five commonly used benchmark datasets: SMD (Server Machine Dataset (Su et al., 2019)), PSM (Polled Server Metrics (Abdulaal et al., 2021)), SWaT (Secure Water Treatment (Mathur & Tippenhauer, 2016)), MSL (Mars Science Laboratory rover), and SMAP (Soil Moisture Active Passive satellite)…”