Table 3 Grid search results for hyperparameter optimization of the MOGKAN model on multi-omics data.

From: Interpretable graph Kolmogorov–Arnold networks for multi-cancer classification and biomarker identification using multi-omics data

Hidden_Dim

Gat_Heads

Hidden_Layers

Dropout_Rate

Learning_rate

L2_penalty

Mean_Accuracy

Mean_F1

2048

4

2

0.2

0.0010

0.0001

0.9618

0.9530

1024

8

2

0.2

0.0010

0.0001

0.9610

0.9514

2048

8

2

0.3

0.0001

0.0001

0.9616

0.9513

1024

4

2

0.3

0.0010

0.0001

0.9604

0.9512

2048

4

2

0.2

0.0001

0.0001

0.9617

0.9512

2048

8

2

0.2

0.0001

0.0001

0.9612

0.9508

1024

8

3

0.2

0.0010

0.0000

0.9608

0.9508

1024

8

2

0.3

0.0010

0.0001

0.9610

0.9507

2048

8

2

0.3

0.0001

0.0000

0.9616

0.9507

2048

8

3

0.2

0.0001

0.0001

0.9608

0.9505