Table 2 Simulation hyperparameters for the proposed model.
S.No | Hyperparameter | Value/Description |
|---|---|---|
1 | Optimizer | Adam |
2 | Learning Rate | 0.001 |
3 | Weight Decay (L2 Regularization) | 0.0005 |
4 | Batch Size | 32 |
5 | Number of Epochs | 50 |
6 | Early Stopping Patience | 10 epochs (monitored on validation loss) |
7 | Dropout Rate | 0.3 |
8 | Activation Function | ReLU |
9 | Number of Hypergraph Layers | 3 |
10 | Number of Capsule Layers | 2 |
11 | Routing Iterations (Capsule) | 3 |
12 | Memory State Dimension (TCMU) | 128 |
13 | Temporal Sequence Length (TCMU) | 10 |
14 | Attention Regularization Weight | 0.1 |
15 | Classification Loss Weight | 1.0 |
16 | Loss Function | Combined (Categorical Cross-Entropy + Attention Regularization) |
17 | Regularization Type | L2 |
18 | Learning Rate Scheduler | Adaptive (ReduceLROnPlateau on validation loss) |
19 | Hardware Used | NVIDIA GPU with 8 GB VRAM (CUDA-accelerated) |
20 | Initialization Scheme | Xavier Normal Initialization |
