Multivariate Time-series Anomaly Detection on SMD 2-1
78AUROCCANDI
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| CANDIalpha=5.0%, Backbone=MLP-based autoencoder2026.04 | 78 | 34.8 | 32.7 | |
| CANDIalpha=0.5%, Backbone=MLP-based autoencoder2026.04 | 72.5 | 31.9 | 26.6 | |
| CANDIalpha=1.0%, Backbone=MLP-based autoencoder2026.04 | 71.1 | 31.4 | 29.2 | |
| M2N2alpha=0.5%, Backbone=MLP-based autoencoder2026.04 | 69.8 | 30.7 | 26.6 | |
| M2N2alpha=1.0%, Backbone=MLP-based autoencoder2026.04 | 69.3 | 30.2 | 29.5 | |
| M2N2alpha=5.0%, Backbone=MLP-based autoencoder2026.04 | 68.3 | 29.6 | 30.9 | |
| Pretrainedalpha=0.5%, Backbone=MLP-based autoencoder2026.04 | 64.8 | 27.5 | 26.5 | |
| Pretrainedalpha=1.0%, Backbone=MLP-based autoencoder2026.04 | 64.8 | 27.5 | 29.6 | |
| Pretrainedalpha=5.0%, Backbone=MLP-based autoencoder2026.04 | 64.8 | 27.5 | 27.3 |