Table 5 Cryptographic parameters for the CKKS scheme at the 128-bit security level, for each kernel and dataset. \(k = 2^{10}\)
From: Private pathological assessment via machine learning and homomorphic encryption
Kernel | ||||
|---|---|---|---|---|
Dataset | Linear | Polynomial | RBF | Sigmoid |
\((N,\log _2 Q)\) | \((N,\log _2 Q)\) | \((N,\log _2 Q)\) | \((N,\log _2 Q)\) | |
Cleveland Heart Disease | (16k, 258) | (32k, 756) | (32k, 804) | (65k, 892) |
Wisconsin Breast Cancer | (16k, 287) | (32k, 644) | (32k, 804) | (65k, 996) |
BreastMNIST | − | − | (64k, 640) | − |
PneumoniaMNIST | − | − | (64k, 640) | − |