Skip to main content
EURASIP Journal on Advances in Signal Processing

Table 2 MSE, PSNR, SSIM for model perturbation defense with privacy budget setting

From: Spears and shields: attacking and defending deep model co-inference in vehicular crowdsensing networks

Privacy Budget \(\epsilon\)

5

50

500

Accuracy

0.6911

0.8871

0.9587

MSE

2589.4388

624.6429

309.7245

PSNR

13.3734

20.1744

22.3587

SSIM

0.2835

0.7463

0.8142

Back to article page