arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2107.06996cs.LGcs.AIcs.NE
收藏

弹性图神经网络

Elastic Graph Neural Networks

Michigan State University(密歇根州立大学);Shandong University(山东大学)

Xiaorui Liu, Wei Jin, Yao Ma, Yaxin Li, Hua Liu, Yiqi Wang, Ming Yan, Jiliang Tang

摘要:

虽然许多现有的图神经网络(GNN)已被证明可以执行基于$\ell_2$的图平滑,从而在全局上强制平滑,但在这项工作中,我们的目标是通过基于$\ell_1$的图平滑进一步增强GNN的局部平滑自适应性。因此,我们介绍了一个家庭的GNN(弹性GNN)的基础上$\ell_1$和$\ell_2$的图形平滑。特别是,我们提出了一个新的和一般的消息传递方案到GNNs。这种消息传递算法不仅对反向传播训练友好,而且在理论收敛保证的情况下实现了所需的平滑特性。在半监督学习任务上的实验表明,所提出的弹性GNN在基准数据集上具有更好的自适应性,并且对图对抗攻击具有显著的鲁棒性。弹性GNN的实现可在\url{https://github.com/lxiaorui/ElasticGNN}上获得。

英文摘要:

While many existing graph neural networks (GNNs) have been proven to perform $\ell_2$-based graph smoothing that enforces smoothness globally, in this work we aim to further enhance the local smoothness adaptivity of GNNs via $\ell_1$-based graph smoothing. As a result, we introduce a family of GNNs (Elastic GNNs) based on $\ell_1$ and $\ell_2$-based graph smoothing. In particular, we propose a novel and general message passing scheme into GNNs. This message passing algorithm is not only friendly to back-propagation training but also achieves the desired smoothing properties with a theoretical convergence guarantee. Experiments on semi-supervised learning tasks demonstrate that the proposed Elastic GNNs obtain better adaptivity on benchmark datasets and are significantly robust to graph adversarial attacks. The implementation of Elastic GNNs is available at \url{https://github.com/lxiaorui/ElasticGNN}.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑