Deep Graph Neural Networks with Shallow Subgraph Samplers

Understand

While Graph Neural Networks (GNNs) are powerful models for learning representations on graphs, most state-of-the-art models do not have significant accuracy gain beyond two to three layers.

• Deep GNNs fundamentally need to address: 1).
• expressivity challenge due to oversmoothing, and 2).
• computation challenge due to neighborhood explosion.