arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1905.06018cs.LGstat.ML
收藏

图神经网络能“上线”吗?预训练与推理分析

Can Graph Neural Networks Go "Online"? An Analysis of Pretraining and Inference

Kiel University(基尔大学);ZBW – Leibniz Information Centre for Economics(莱布尼茨经济信息中心);University of Essex(埃塞克斯大学)

Lukas Galke, Iacopo Vagliano, Ansgar Scherp

摘要:

实际应用中的大规模图数据往往不是静态的,而是动态的。例如,新的节点和边随着时间出现。当前的图卷积方法是有前途的,特别是当所有图的节点和边在训练期间可用时。当在训练后插入看不见的节点和边时,尚未评估向上训练还是从头开始重新训练是优选的。我们构建了一个实验设置,在该设置中,我们在训练后插入以前看不见的节点和边,并进行有限数量的推理时期。在这个设置中,我们比较了适应预训练的图神经网络和从头开始重新训练。我们的研究结果表明,预训练的模型在看不见的节点上产生了高准确性分数,并且预训练优于从头开始重新训练。我们的实验代表了评估和开发真正的图神经网络在线变体的第一步。

英文摘要:

Large-scale graph data in real-world applications is often not static but dynamic, i. e., new nodes and edges appear over time. Current graph convolution approaches are promising, especially, when all the graph's nodes and edges are available during training. When unseen nodes and edges are inserted after training, it is not yet evaluated whether up-training or re-training from scratch is preferable. We construct an experimental setup, in which we insert previously unseen nodes and edges after training and conduct a limited amount of inference epochs. In this setup, we compare adapting pretrained graph neural networks against retraining from scratch. Our results show that pretrained models yield high accuracy scores on the unseen nodes and that pretraining is preferable over retraining from scratch. Our experiments represent a first step to evaluate and develop truly online variants of graph neural networks.

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。

↑