Node-wise Filtering in Graph Neural Networks: A Mixture of Experts Approach
This addresses a key limitation in GNNs for node classification tasks on real-world graphs with complex structural patterns, representing an incremental advancement.
The paper tackles the problem of suboptimal performance in Graph Neural Networks (GNNs) when using a single global filter for graphs with mixed homophilic and heterophilic patterns, by introducing Node-MoE, a framework that adaptively selects filters per node, achieving improved results on both types of graphs.
Graph Neural Networks (GNNs) have proven to be highly effective for node classification tasks across diverse graph structural patterns. Traditionally, GNNs employ a uniform global filter, typically a low-pass filter for homophilic graphs and a high-pass filter for heterophilic graphs. However, real-world graphs often exhibit a complex mix of homophilic and heterophilic patterns, rendering a single global filter approach suboptimal. In this work, we theoretically demonstrate that a global filter optimized for one pattern can adversely affect performance on nodes with differing patterns. To address this, we introduce a novel GNN framework Node-MoE that utilizes a mixture of experts to adaptively select the appropriate filters for different nodes. Extensive experiments demonstrate the effectiveness of Node-MoE on both homophilic and heterophilic graphs.