Table 1.

Search ranges for hyperparameters of MoGAT.

Hyperparameter	Search range
$\kappa$	1, 2, 3, 4, 5, 6
$\tau$	1, 2, 3, 4, 5, 6
$d_{\mathcal{G}}$	100, 150, 200, 250, 300, 350, 400
Dropout rate	0.1, 0.2, 0.3, 0.4, 0.5
Learning rate	0.1, 0.01, 0.001, 0.0001

The optimal values are highlighted in boldface. $\kappa$ and $\tau$ denote the number of node embedding and graph embedding updates, respectively. In addition, $d_{\mathcal{G}}$ denotes dimension of the embeddings.