Table 2 VAE hyperparameter settings
From: Deep reinforcement learning for data-efficient weakly supervised business process anomaly detection
| Â | Encoder | Decoder |
|---|---|---|
Input layer size | Trace length * Number of activities | 5 |
No. of hidden layers | 2 | 2 |
1st hidden layer size | Input layer size/2 | Input layer size/4 |
2nd hidden layer size | Input layer size/4 | Input layer size/2 |
Output layer size | 5 | Encoder input layer size |
Activation functions | Relu | Relu |