Table 2 Training parameters of machine learning models.
From: Machine learning-based detection and mitigation of cyberattacks in adaptive cruise control systems
Model | Training parameters | |||
|---|---|---|---|---|
RF | No of trees | 100 | Maximum Depth | 10 |
Minimum Sample Split | 2 | Split Criteria | Gini Index | |
SVM | Regularization | 100 | Feature Scaling | Minmax |
Epsilon | 0.1 | Kernel | rbf | |
TDNN | No of Training Epochs | 20 | No of Hidden Units | 15 |
Window Length | 10 | Feature Scaling | Minmax | |
Learning Algorithm | SGD | Hidden Layer Activation | ReLU | |
FNN | No of Training Epochs | 20 | No of Hidden Units | 15 |
No of Hidden Layers | 1 | Feature Scaling | Minmax | |
Learning Algorithm | SGD | Hidden Layer Activation | ReLU | |
SGD: | Stochastic Gradient Descent | ReLU: | Rectified Linear Activation | |
