Auxiliary learning induced graph convolutional networks

Gengchen Duan, Taisong Jin, Rongrong Ji, Ling Shao, Baochang Zhang, Feiyue Huang, YONGJIAN WU

Keywords: graph neural networks, meta learning, auxiliary learning

Abstract: In this article, we propose a novel auxiliary learning induced graph convolutional network in a multi-task fashion. Specifically, both the link prediction and pseudo label generation are used as two auxiliary tasks to complement the primary task of node classification. Those two auxiliary tasks are jointly trained with the primary node classification task via the graph meta-learning strategy. The experimental results demonstrate that the proposed method consistently and significantly outperforms the existing methods and achieves state-of-the-art node classification results on the benchmark citation network datasets.

Code Of Conduct: I certify that all co-authors of this work have read and commit to adhering to the NeurIPS Statement on Ethics, Fairness, Inclusivity, and Code of Conduct.