Table 8 Comparison experiments in noisy environments.
From: A framework for hardware trojan detection based on contrastive learning
| Â | Method | TPR % | TNR % | FPR % | FNR % | Accuracy % | Precision % | F1 Score |
|---|---|---|---|---|---|---|---|---|
T100-2100 | DCNN-ResNet | 30.00 | 70.00 | 0.30 | 0.70 | 50.00 | 50.00 | 0.38 |
Our Proposal | 34.00 | 100.00 | 0.00 | 0.66 | 67.00 | 100.00 | 0.51 | |
T2100-100 | DCNN-ResNet | 60.00 | 32.00 | 0.68 | 0.40 | 46.00 | 46.88 | 0.53 |
Our Proposal | 56.00 | 64.00 | 0.36 | 0.44 | 60.00 | 60.87 | 0.58 |
