Table 2 Training parameters.

From: Fusing temporal and structural information via subgraph sampling and multi-head attention for information cascade prediction

Module

Parameter

Value

MH-GAT

Input/Output Dimensions

32

Attention Heads

2

Layers

2

Bi-GRU

Hidden Dimensions

64

Layers

2

SENet

Reduction Ratio

16 → 8

MLP

Layers

64 → 32 → 1

Optimizer

Learning Rate

0.002

Back to article page