A Robust Stacking Framework for Training Deep Graph Models with Multifaceted Node Features
This work addresses the problem of effectively combining graph and non-graph models for practitioners dealing with multifaceted node features, representing an incremental improvement over existing hybrid strategies.
The authors tackled the challenge of integrating non-neural models for tabular or text data into graph neural networks (GNNs) by proposing a robust stacking framework that fuses graph-aware propagation with arbitrary IID models, achieving comparable or superior performance across various graph datasets.
Graph Neural Networks (GNNs) with numerical node features and graph structure as inputs have demonstrated superior performance on various supervised learning tasks with graph data. However the numerical node features utilized by GNNs are commonly extracted from raw data which is of text or tabular (numeric/categorical) type in most real-world applications. The best models for such data types in most standard supervised learning settings with IID (non-graph) data are not simple neural network layers and thus are not easily incorporated into a GNN. Here we propose a robust stacking framework that fuses graph-aware propagation with arbitrary models intended for IID data, which are ensembled and stacked in multiple layers. Our layer-wise framework leverages bagging and stacking strategies to enjoy strong generalization, in a manner which effectively mitigates label leakage and overfitting. Across a variety of graph datasets with tabular/text node features, our method achieves comparable or superior performance relative to both tabular/text and graph neural network models, as well as existing state-of-the-art hybrid strategies that combine the two.