Table 3 Experimental parameters.
Parameter | Value |
|---|---|
Number of Hospitals | 10 |
Number of Patients per Hospital | 1000 |
Training Rounds | 100 |
Batch Size | 64 |
Learning Rate | 0.001 |
Optimizer | Adam |
Encryption Scheme | PHC |
Encryption Key Size | 2048 bits |
Federated Learning Framework | TensorFlow |
Local Epochs | 5 |
Neural Network Architecture | Residual Learning based DBN |
Number of Hidden Layers (RDBN) | 5 |
Hidden Units per Layer (RDBN) | 256 |
Activation Function | ReLU |
Dropout Rate | 0.5 |
Loss Function | Cross-Entropy |
Noise Level for Differential Privacy | 1.0 |
Q-Learning Learning Rate | 0.1 |
Q-Learning Discount Factor | 0.9 |
ISD-k-ADP Anonymity Level (k) | 5 |
ISD-k-ADP Perturbation Level | Medium |
CCM-PPAMP Chaotic Map Dimension | 3 |
CCM-PPAMP Authentication Key Size | 256 bits |
IMFPM Piecewise Mechanism Segments | 10 |
IMFPM Regularization Parameter | 0.01 |
Communication Rounds (FL) | 50 |
Model Aggregation Method (FL) | Federated Averaging |
