Graph ensemble neural network

Rui Duan, Chungang Yan, Junli Wang, Changjun Jiang

Abstract: Highlights•Graph ensemble learning ignores the hierarchical nature of GNNs.•Graph ensemble learning fail to promote many GNNs to perform well under heterophily.•Multiple deep ensembles throughout the neighbor aggregation to fuse multiple graphs.•Our model achieve state-of-the-art performance on homophily and heterophily graphs.