Timeseries Anomaly Detection using Temporal Hierarchical One-Class Network

By Lifeng Shen, Zhuocong Li, James T. Kwok. Neural Information Processing Systems, 2020

425 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/fc36d50be4352afada233f55a59f0dca0a7a826b

How researchers see this paper

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

1. Wojciech Samek
2. Grgoire Montavon
3. Mingsheng Long
4. Jianmin Wang
5. Zhijian Xu
6. K. Müller
7. Haixu Wu
8. Lukas Ruff

Citations in the corpus

• Deep Learning for Time Series Anomaly Detection: A Survey (2022, unknown): “THOC [155] represents a self-supervised temporal hierarchical one-class network, which consists of a multilayer dilated RNN and a hierarchical SVDD [164].”
• ADBench: Anomaly Detection Benchmark (2022, background): “AD algorithms aim to identify data instances that deviate significantly from the majority of data objects [59, 139, 146, 160], and numerous methods have been developed in the last few decades [3, 85, 102, 103, 129, 156, 172, 198].”
• Deep Isolation Forest for Anomaly Detection (2022, methodology): “Following [34], [37], they directly handle time-series datasets by treating each timestamp as a data object.”
• Multivariate Time Series Anomaly Detection and Interpretation using Hierarchical Inter-Metric and Temporal Embedding (2021, methodology): “However, it relies on the covariance among different metrics, and cannot well learn the complex and nonlinear interactions among metrics [27].”
• Anomaly Transformer: Time Series Anomaly Detection with Association Discrepancy (2021, methodology): “Baselines We extensively compare our model with 18 baselines, including the reconstructionbased models: InterFusion (2021), BeatGAN (2019), OmniAnomaly (2019), LSTM-VAE (2018); the density-estimation models: DAGMM (2018), MPPCACD (2017), LOF (2000); the clusteringbased methods: ITAD (2020), THOC (20”
• FITS: Modeling Time Series with 10k Parameters (2023, methodology): “Following the previous works (Xu et al., 2022; Shen et al., 2020; Wu et al., 2023), we use Precision, Recall, and F1-score as metrics.”
• A Unifying Review of Deep and Shallow Anomaly Detection (2020, background): “Further variants of deep one-class classification include multimodal [145] or time-series extensions [338] and methods that employ adversarial learning [138], [141], [339] or transfer learning [139], [142].”
• Deep Learning for Anomaly Detection in Time-Series Data: Review, Analysis, and Guidelines (2021, methodology): “…categorized as follows : • AE: DAGMM [103], MSCRED [51], OmniAnomaly [68] • VAE: LSTM-VAE [112], USAD [72] • GAN: MAD-GAN [59] • RNN: THOC [60] • Transformer: GTA [109] • GNN: GTA [109], GDN [108] Table 6 shows the anomaly detection accuracy in terms of Precision, Recall, and F1-score of…”