arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2201.00115cs.SEcs.LG
收藏

关于图神经网络的分析

Toward the Analysis of Graph Neural Networks

University of Melbourne(墨尔本大学);Hanoi University of Science and Technology(河内科技大学);George Mason University(乔治梅森大学)

Thanh-Dat Nguyen, Thanh Le-Cong, ThanhVu H. Nguyen, Xuan-Bach D. Le, Quyet-Thang Huynh

摘要:

图神经网络(GNN)最近已经成为图结构数据的鲁棒框架。它们已被应用于知识图谱分析、社交网络推荐,甚至Covid 19检测和疫苗开发等诸多问题。然而,与其他深度神经网络(如前馈神经网络(FFNN))不同,很少有分析(如验证和属性推断)存在,这可能是由于GNN的动态行为,GNN可以将任意图形作为输入,而FFNN仅将固定大小的数值向量作为输入。本文提出了一种将GNN转换为FFNN并重用现有FFNN分析的方法。我们讨论了各种设计,以确保转换的可扩展性和准确性。我们说明我们的方法上的一个研究案例的节点分类。我们相信,我们的方法为理解和分析GNN开辟了新的研究方向。

英文摘要:

Graph Neural Networks (GNNs) have recently emerged as a robust framework for graph-structured data. They have been applied to many problems such as knowledge graph analysis, social networks recommendation, and even Covid19 detection and vaccine developments. However, unlike other deep neural networks such as Feed Forward Neural Networks (FFNNs), few analyses such as verification and property inferences exist, potentially due to dynamic behaviors of GNNs, which can take arbitrary graphs as input, whereas FFNNs which only take fixed size numerical vectors as inputs. This paper proposes an approach to analyze GNNs by converting them into FFNNs and reusing existing FFNNs analyses. We discuss various designs to ensure the scalability and accuracy of the conversions. We illustrate our method on a study case of node classification. We believe that our approach opens new research directions for understanding and analyzing GNNs.

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。

↑