Realistic graphs contain both rich self-features of nodes and informative
structures of neighborhoods, jointly handled by a GNN in the typical setup. We
propose to decouple the two modalities by mixture of weak and strong experts
(Mowst), where the weak expert is a light-weight Multi-layer Perceptron (MLP),
and the strong expert is an off-the-shelf Graph Neural Network (GNN). To adapt
the experts' collaboration to different target nodes, we propose a "confidence"
mechanism based on the dispersion of the weak expert's prediction logits. The
strong expert is conditionally activated when either the node's classification
relies on neighborhood information, or the weak expert has low model quality.
We reveal interesting training dynamics by analyzing the influence of the
confidence function on loss: our training algorithm encourages the
specialization of each expert by effectively generating soft splitting of the
graph. In addition, our "confidence" design imposes a desirable bias toward the
strong expert to benefit from GNN's better generalization capability. Mowst is
easy to optimize and achieves strong expressive power, with a computation cost
comparable to a single GNN. Empirically, Mowst shows significant accuracy
improvement on 6 standard node classification benchmarks (including both
homophilous and heterophilous graphs).

通过混合弱和强专家的方式，使用基于多层感知器（MLP）的轻量级专家来处理节点的自我特征和领域的信息结构，其中根据弱专家预测的置信度来条件激活强专家，以实现节点分类的准确性提升。