P-DROP: Poisson-Based Dropout for Graph Neural Networks
This addresses over-smoothing in GNNs for graph learning tasks, but it is incremental as it builds on existing dropout techniques.
The paper tackles the over-smoothing problem in Graph Neural Networks by proposing a Poisson-based node selection strategy for stochastic updates, resulting in competitive or improved accuracy on benchmarks like Cora, Citeseer, and Pubmed compared to traditional dropout methods.
Over-smoothing remains a major challenge in Graph Neural Networks (GNNs), where repeated message passing causes node representations to converge and lose discriminative power. To address this, we propose a novel node selection strategy based on Poisson processes, introducing stochastic but structure-aware updates. Specifically, we equip each node with an independent Poisson clock, enabling asynchronous and localized updates that preserve structural diversity. We explore two applications of this strategy: as a replacement for dropout-based regularization and as a dynamic subgraph training scheme. Experimental results on standard benchmarks (Cora, Citeseer, Pubmed) demonstrate that our Poisson-based method yields competitive or improved accuracy compared to traditional Dropout, DropEdge, and DropNode approaches, particularly in later training stages.