Efficient ensembles of graph neural networks

What is it about?

Graph Neural Networks (GNNs) have attracted a lot of attention in recent years by achieving state-of-the-art performance in graph processing tasks. This work proposes techniques that simultaneously improve both accuracy and inference efficiency of GNNs.

Featured Image

Photo by JJ Ying on Unsplash

Photo by JJ Ying on Unsplash

Why is it important?

Ensembles improve the accuracy and robustness of Graph Neural Networks (GNNs), but suffer from high latency and storage requirements. To address this challenge, we propose GNN Ensembles with Error Node Isolation (GEENI). The key concept in GEENI is to identify nodes that are likely to be incorrectly classified (error nodes) and suppress their outgoing messages, leading to simultaneous accuracy and efficiency improvements. GEENI also enables aggressive approximations of the constituent models in the ensemble while maintaining accuracy. To improve the efficacy of GEENI, we propose techniques for diverse ensemble creation and accurate error node identification. Our experiments establish that GEENI models are simultaneously up to 4.6% (3.8%) more accurate and up to 2.8X (5.7X) faster compared to non-ensemble (conventional ensemble) GNN models.