GenIAS: Generator for Instantiating Anomalies in time Series
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866918324073398272 |
|---|---|
| author | Darban, Zahra Zamanzadeh Wang, Qizhou Webb, Geoffrey I. Pan, Shirui Aggarwal, Charu C. Salehi, Mahsa |
| author_facet | Darban, Zahra Zamanzadeh Wang, Qizhou Webb, Geoffrey I. Pan, Shirui Aggarwal, Charu C. Salehi, Mahsa |
| contents | Synthetic anomaly injection is a recent and promising approach for time series anomaly detection (TSAD), but existing methods rely on ad hoc, hand-crafted strategies applied to raw time series that fail to capture diverse and complex anomalous patterns, particularly in multivariate settings. We propose a synthetic anomaly generation method named Generator for Instantiating Anomalies in Time Series (GenIAS), which generates realistic and diverse anomalies via a novel learnable perturbation in the latent space of a variational autoencoder. This enables abnormal patterns to be injected across different temporal segments at varying scales based on variational reparameterization. To generate anomalies that align with normal patterns while remaining distinguishable, we introduce a learning strategy that jointly learns the perturbation scale and compact latent representations via a tunable prior, which improves the distinguishability of generated anomalies, as supported by our theoretical analysis. Extensive experiments show that GenIAS produces more diverse and realistic anomalies, and that detection models trained with these anomalies outperform 17 baseline methods on 9 popular TSAD benchmarks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_08262 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | GenIAS: Generator for Instantiating Anomalies in time Series Darban, Zahra Zamanzadeh Wang, Qizhou Webb, Geoffrey I. Pan, Shirui Aggarwal, Charu C. Salehi, Mahsa Machine Learning Synthetic anomaly injection is a recent and promising approach for time series anomaly detection (TSAD), but existing methods rely on ad hoc, hand-crafted strategies applied to raw time series that fail to capture diverse and complex anomalous patterns, particularly in multivariate settings. We propose a synthetic anomaly generation method named Generator for Instantiating Anomalies in Time Series (GenIAS), which generates realistic and diverse anomalies via a novel learnable perturbation in the latent space of a variational autoencoder. This enables abnormal patterns to be injected across different temporal segments at varying scales based on variational reparameterization. To generate anomalies that align with normal patterns while remaining distinguishable, we introduce a learning strategy that jointly learns the perturbation scale and compact latent representations via a tunable prior, which improves the distinguishability of generated anomalies, as supported by our theoretical analysis. Extensive experiments show that GenIAS produces more diverse and realistic anomalies, and that detection models trained with these anomalies outperform 17 baseline methods on 9 popular TSAD benchmarks. |
| title | GenIAS: Generator for Instantiating Anomalies in time Series |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2502.08262 |