Multivariate Time-series Anomaly Detection on SMD 3-2
0.717AUROCCANDI
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| CANDIalpha=0.5%, Backbone=MLP-based autoencoder2026.04 | 0.717 | 0.199 | 0.017 | |
| CANDIalpha=1.0%, Backbone=MLP-based autoencoder2026.04 | 0.717 | 0.199 | 0.03 | |
| CANDIalpha=5.0%, Backbone=MLP-based autoencoder2026.04 | 0.717 | 0.199 | 0.262 | |
| M2N2alpha=5.0%, Backbone=MLP-based autoencoder2026.04 | 0.64 | 0.188 | 0.194 | |
| M2N2alpha=1.0%, Backbone=MLP-based autoencoder2026.04 | 0.632 | 0.179 | 0.03 | |
| M2N2alpha=0.5%, Backbone=MLP-based autoencoder2026.04 | 0.573 | 0.174 | 0.017 | |
| Pretrainedalpha=0.5%, Backbone=MLP-based autoencoder2026.04 | 0.451 | 0.159 | 0.017 | |
| Pretrainedalpha=1.0%, Backbone=MLP-based autoencoder2026.04 | 0.451 | 0.159 | 0.031 | |
| Pretrainedalpha=5.0%, Backbone=MLP-based autoencoder2026.04 | 0.451 | 0.159 | 0.247 |