Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2605.27486 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913165373079552 |
|---|---|
| author | Nosrati, Khayyam Uray, Martin Messineo, Saverio Sassnick, Olaf Huber, Stefan |
| author_facet | Nosrati, Khayyam Uray, Martin Messineo, Saverio Sassnick, Olaf Huber, Stefan |
| contents | Federated learning (FL) has broadened the horizon for multivariate time series anomaly detection (MTSAD). However, benchmarking such anomaly detection methods within FL paradigm poses data-centric challenges. The existing datasets do not counteract these challenges since they do not simultaneously provide sufficient scale, accurate labels, and freedom from common flaws. In addition, the role of cyclic process behavior, which is common in discrete industrial automation, remains underexplored for MTSAD for the current state of research. This paper aims to shed more light on the literature and address these gaps by introducing a dataset designed with cyclic dynamics arising from the repetitive nature of discrete automation processes and evaluates selected MTSAD methods on both the proposed dataset and a public benchmark dataset. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_27486 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Federated Learning for Multivariate Time Series Anomaly Detection in Industrial Automation Nosrati, Khayyam Uray, Martin Messineo, Saverio Sassnick, Olaf Huber, Stefan Machine Learning Federated learning (FL) has broadened the horizon for multivariate time series anomaly detection (MTSAD). However, benchmarking such anomaly detection methods within FL paradigm poses data-centric challenges. The existing datasets do not counteract these challenges since they do not simultaneously provide sufficient scale, accurate labels, and freedom from common flaws. In addition, the role of cyclic process behavior, which is common in discrete industrial automation, remains underexplored for MTSAD for the current state of research. This paper aims to shed more light on the literature and address these gaps by introducing a dataset designed with cyclic dynamics arising from the repetitive nature of discrete automation processes and evaluates selected MTSAD methods on both the proposed dataset and a public benchmark dataset. |
| title | Federated Learning for Multivariate Time Series Anomaly Detection in Industrial Automation |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2605.27486 |