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 1806.00770cs.LGcs.AIstat.ML
收藏

双本原图卷积网络

Dual-Primal Graph Convolutional Networks

USI(瑞士意大利语区大学);Fabula AI(Fabula AI公司);TU Munich(慕尼黑工业大学);Tel-Aviv University(特拉维夫大学);Intel Perceptual Computing(英特尔感知计算部门);Imperial College London(帝国理工学院)

Federico Monti, Oleksandr Shchur, Aleksandar Bojchevski, Or Litany, Stephan Günnemann, Michael M. Bronstein

摘要:

近年来,人们对开发用于图等非欧几里得结构化数据的深度学习方法的兴趣激增。在本文中,我们提出了Dual-Primal Graph CNN,这是一种图形卷积架构,可以在图形及其对偶上交替进行类似卷积的操作。我们的方法允许学习顶点和边的功能,并概括了以前的图形注意力(GAT)模型。我们提供了广泛的实验验证,展示了在已建立的图形基准测试上测试的各种任务的最先进的结果,包括CORA和Citeseer引文网络以及MovieLens,Flixter,豆瓣和雅虎音乐图形引导推荐系统。

英文摘要:

In recent years, there has been a surge of interest in developing deep learning methods for non-Euclidean structured data such as graphs. In this paper, we propose Dual-Primal Graph CNN, a graph convolutional architecture that alternates convolution-like operations on the graph and its dual. Our approach allows to learn both vertex- and edge features and generalizes the previous graph attention (GAT) model. We provide extensive experimental validation showing state-of-the-art results on a variety of tasks tested on established graph benchmarks, including CORA and Citeseer citation networks as well as MovieLens, Flixter, Douban and Yahoo Music graph-guided recommender systems.

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。

↑