Details
| Original language | English |
|---|---|
| Article number | 26704 |
| Journal | Scientific reports |
| Volume | 14 |
| Issue number | 1 |
| Publication status | Published - 4 Nov 2024 |
Abstract
In Federated Learning, Anomaly Detection poses significant challenges due to the decentralized nature of data, especially under Non-IID distributions. This study proposes a federated threshold calculation method that aggregates summary statistics from normal and anomalous data across clients to create a global threshold for Anomaly Detection with federated Autoencoders, enhancing detection accuracy and robustness while ensuring privacy. Extensive experiments on datasets, including Credit Card Fraud Detection, Shuttle, and Covertype, show that our approach consistently outperforms existing federated and local threshold calculation methods. These findings highlight the potential of summary statistics in improving federated Anomaly Detection under Non-IID conditions.
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: Scientific reports, Vol. 14, No. 1, 26704, 04.11.2024.
Research output: Contribution to journal › Article › Research › peer review
}
TY - JOUR
T1 - Enhanced federated anomaly detection through autoencoders using summary statistics-based thresholding
AU - Laridi, Sofiane
AU - Palmer, Gregory
AU - Tam, Kam Ming Mark
N1 - Publisher Copyright: © The Author(s) 2024.
PY - 2024/11/4
Y1 - 2024/11/4
N2 - In Federated Learning, Anomaly Detection poses significant challenges due to the decentralized nature of data, especially under Non-IID distributions. This study proposes a federated threshold calculation method that aggregates summary statistics from normal and anomalous data across clients to create a global threshold for Anomaly Detection with federated Autoencoders, enhancing detection accuracy and robustness while ensuring privacy. Extensive experiments on datasets, including Credit Card Fraud Detection, Shuttle, and Covertype, show that our approach consistently outperforms existing federated and local threshold calculation methods. These findings highlight the potential of summary statistics in improving federated Anomaly Detection under Non-IID conditions.
AB - In Federated Learning, Anomaly Detection poses significant challenges due to the decentralized nature of data, especially under Non-IID distributions. This study proposes a federated threshold calculation method that aggregates summary statistics from normal and anomalous data across clients to create a global threshold for Anomaly Detection with federated Autoencoders, enhancing detection accuracy and robustness while ensuring privacy. Extensive experiments on datasets, including Credit Card Fraud Detection, Shuttle, and Covertype, show that our approach consistently outperforms existing federated and local threshold calculation methods. These findings highlight the potential of summary statistics in improving federated Anomaly Detection under Non-IID conditions.
KW - Anomaly Detection
KW - Autoencoders
KW - Federated Learning
UR - http://www.scopus.com/inward/record.url?scp=85208516884&partnerID=8YFLogxK
U2 - 10.1038/s41598-024-76961-2
DO - 10.1038/s41598-024-76961-2
M3 - Article
C2 - 39496691
AN - SCOPUS:85208516884
VL - 14
JO - Scientific reports
JF - Scientific reports
SN - 2045-2322
IS - 1
M1 - 26704
ER -