Graph Neural Networks (GNNs), due to their capability to learn complex relations (edges) among attributed objects (nodes) within graph datasets, have already been widely used in various graph mining tasks. Considerable efforts have been devoted to improving GNN learning through designing new architectures and/or loss objectives. In this paper, we introduce a novel GNN learning framework, called AKE-GNN (Adaptive-Knowledge-Exchange GNN), which adaptively exchanges diverse knowledge learned from multiple graph views generated by graph augmentations. Specifically, AKE-GNN iteratively exchanges redundant channels in the weight matrix of one GNN by informative channels of another GNN in a layer-wise manner. Furthermore, existing GNN models can be seamlessly incorporated into our framework. Extensive experiments on node classification, graph classification, and edge prediction demonstrate the effectiveness of AKE-GNN. In particular, we conduct a series of experiments on 15 public benchmarks, 8 popular GNN models, and 3 graph tasks -- node classification, graph classification, and edge prediction -- and show that AKE-GNN consistently outperforms existing popular GNN models and even their ensembles. On the Cora semi-supervised node classification dataset, our framework achieves new state-of-the-art results. Extensive ablation studies and analyses on knowledge exchange methods also verify the effectiveness of AKE-GNN.

翻译：由于有能力在图表数据集中学习被归属对象(节点)之间的复杂关系(格外网),因此在各种图表采矿任务中已广泛使用。已经作出大量努力,通过设计新的架构和/或损失目标来改进GNN学习。在本文件中,我们引入了一个新的GNN学习框架,称为Adaption-Knewledge-Exchange GNNN(Adapide-Knew-Knewledge-Exchange GNN)(GNNN)(GNN)(GN)(GNN)(GN)(GNN)(G)(GNN)(GNN)(G)(GNN(G)(GNNN)(G)(GNN)(G)(GNN)(G)(GNN(G)(GNN)(GNN)(G) (GNN) (G) (G) (GNN(NNN) (G) (G) (GNNN) (G) (G) (G) (GNNN) (G(G) (G) (G(NNNN) (G) (G) (G) (GNN) (GN) (GN) (GN) (G(G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G(G) (G) (G) (G) (G) (GNNNNNNNN) (G(G(G) (G) (G(G(G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (G) (N) ((