metrics such as the minimum (the smallest value in the dataset), maximum (the largest value), median (the middle value when arranged in increasing order), first quartile (Q1: the middle value between the minimum and the median), and third quartile (Q3: the middle value between the median and the maximum). Analyzing the box plot as illustrated in Figures 10 and 11, we have added a few things to reduce overfitting: Ñ Early stopping: By stopping training if validation loss does not improve for 20 epochs, we prevent the model from overfitting to the training data. If the validation loss is no longer improving, continued training is unlikely to generalize better to new data. Ñ Restore best weights: By restoring weights from the epoch with the best validation loss, we “roll back” the model to the point before overfitting started to occur. This gives us the model that generalizes best to new data. ME |AI/ML 6 5 4 3 2 1 0 0 25 50 75 100 125 150 175 200 Training loss Epoch Mean loss modell Loss range modell Mean loss model with TL Loss range model with TL 5 4 3 2 1 0 0 25 50 75 100 125 150 175 200 Training loss Epoch Mean loss modell Loss range modell Mean loss model with TL Loss range model with TL M L M L 5 4 3 2 1 0 0 25 50 75 100 125 150 175 200 Validation loss Epoch Mean modelled Loss range modell Mean model with TL Loss range model with TL 6 5 4 3 2 1 0 0 25 50 75 100 125 150 175 200 Validation loss Epoch Mean modell Loss range modell Mean model with TL Loss range model with TL M lloss de range Mean lloss s model del th M lloss L range an lloss s m el L del th 7 6 5 4 3 2 1 0 25 50 75 100 125 150 175 200 Training loss Epoch Mean loss model Loss range model Mean model with TL Loss range model with TL 7 6 5 4 3 2 1 0 25 50 75 100 125 150 175 200 Training loss Epoch Mean modeled Loss range modeled Mean model with TL Loss range model with TL M L de M lloss del h L range M lloss L M lloss del th L range 7 6 5 4 3 2 1 0 25 50 75 100 125 150 175 200 Validation loss Epoch Mean loss model Loss range model Mean loss model with TL Loss range model with TL 7 6 5 4 3 2 1 0 25 50 75 100 125 150 175 200 Validation loss Epoch Mean modeled Loss range model Mean loss model with TL Loss range model with TL M L M h L M lloss L M m l w L Figure 8. Comparative analysis of mean loss and range with and without the implementation of transfer learning for: (a) Encoder model applied to the impact test dataset (b) Encoder model applied to the PLB test dataset (c) MLP model applied to the impact test dataset and (d) MLP model applied to the PLB test dataset. 80 M A T E R I A L S E V A L U A T I O N • J U L Y 2 0 2 3 2307 ME July dup.indd 80 6/19/23 3:41 PM

Ñ Patience: The patience value of 20 epochs means we are willing to tolerate a fair number of epochs without improve- ment before stopping training. This avoids stopping too early and allows temporary plateaus in validation loss, but ulti- mately stops before severe overfitting occurs. In the Impact dataset, the CNN and MLP models, with and without transfer learning, achieved comparative perfor- mance in terms of accuracy, precision, and recall, with slight enhancements observed in models using transfer learning. Conversely, FCNN underperformed, showing negligible improvement from transfer learning unlike CNN and MLP, which recorded accuracies above 0.8, FCNN yielded a mere 0.2. Transfer learning substantially increased ResNet’s performance variance regarding recall, precision, and accuracy. Inception showed a similar trend to CNN and MLP, where transfer learning resulted in minor enhancements. The Encoder model Training loss Epoch 20 15 10 5 0 0 100 200 300 400 500 Mean loss model Loss range modell Mean loss model with TL Loss range model with TL L de 20 15 10 5 0 0 100 200 300 400 500 Mean loss model Loss range modell Mean model with TL Loss range model with TL Validation loss Epoch e lloss del th del th Training loss Epoch 14 12 10 8 6 4 2 0 0 100 200 300 400 500 Mean loss modell Loss range modell Mean loss model with TL Loss range model with TL M L M L 14 12 10 8 6 4 2 0 0 100 200 300 400 500 Mean loss Loss range modelled Mean model with TL Loss range model with TL Validation loss Epoch an mmodel L M lloss del th L del th Training loss Epoch 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0 25 50 75 100 125 150 175 200 Mean modell Loss range modell Mean model with TL Loss range model with TL M lloss de L de an lloss s m el L del th 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0 25 50 75 100 125 150 175 200 Mean modell Loss range modell Mean loss model with TL Loss range model with TL Validation loss Epoch M lloss de L de M l w L del th Training loss Epoch 4 3 2 1 0 0 25 50 75 100 125 150 175 200 Mean modelled Loss range modelled Mean model with TL Loss range model with TL M lloss L an lloss s m el L del th 4 3 2 1 0 0 25 50 75 100 125 150 175 200 Mean modelled Loss range modelled Mean loss model with TL Loss range model with TL Validation loss Epoch M lloss L M L del th Figure 9. Comparative analysis of mean loss and range with and without the implementation of transfer learning for: (a) Inception model applied to the impact test dataset (b) Inception model applied to the PLB test dataset (c) ResNet model applied to the impact test dataset and (d) ResNet model applied to the PLB test dataset. J U L Y 2 0 2 3 • M A T E R I A L S E V A L U A T I O N 81 2307 ME July dup.indd 81 6/19/23 3:41 PM