Enhanced Graph Neural Networks using K-Hop Gaussian Diffusion
For graph neural network practitioners, this method improves performance on graphs with noisy edges or complex structures, though it is an incremental improvement over existing diffusion kernels.
The paper introduces a K-Hop Gaussian diffusion kernel that preprocesses graph data to balance local and global information propagation, outperforming traditional GNNs and existing diffusion methods on noisy or structurally complex graphs.
Most graph neural network (GNN) cores rely on graph convolutions, typically implemented as message passing between direct (single-hop) neighbors. In many real-world graphs, edges can be noisy or poorly defined, limiting information propagation to local neighborhoods. Existing diffusion kernels, such as Personalized PageRank (PPR) and Heat Kernel, alleviate this issue through global propagation, but still struggle with complex local structures and distant node noise. To address these limitations, we propose a K-Hop Gaussian (KHG) diffusion kernel as a preprocessing module for graph data. KHG introduces multi-hop diffusion with Gaussian weighting for remote nodes, balancing local and global information propagation before applying standard GNNs. Experiments on multiple benchmark datasets demonstrate that KHG significantly outperforms traditional message-passing GNNs, as well as PPR and Heat Kernel diffusion, particularly in noisy or structurally complex graphs.