arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2103.00137cs.LGcs.AI
收藏

基于图神经网络的元学习方法及应用

Meta-Learning with Graph Neural Networks: Methods and Applications

Max Planck Institute for Software Systems(马克斯·普朗克软件系统研究所);Northwestern University(西北大学);IBM T. J. Watson Research Center(IBM T. J. 沃森研究中心)

Debmalya Mandal, Sourav Medya, Brian Uzzi, Charu Aggarwal

摘要:

图神经网络(GNN)是深度神经网络在图数据上的推广,已广泛应用于从药物发现到推荐系统的各个领域。然而,当可用样本很少时,这些应用程序上的GNN是有限的。元学习一直是解决机器学习中样本缺乏的重要框架,近年来,研究人员开始将元学习应用于GNN。在这项工作中,我们提供了一个全面的调查不同的元学习方法,涉及GNN对各种图形问题显示使用这两种方法一起的力量。我们根据建议的架构,共享表示和应用程序对文献进行分类。最后,我们讨论了几个令人兴奋的未来研究方向和开放的问题。

英文摘要:

Graph Neural Networks (GNNs), a generalization of deep neural networks on graph data have been widely used in various domains, ranging from drug discovery to recommender systems. However, GNNs on such applications are limited when there are few available samples. Meta-learning has been an important framework to address the lack of samples in machine learning, and in recent years, researchers have started to apply meta-learning to GNNs. In this work, we provide a comprehensive survey of different meta-learning approaches involving GNNs on various graph problems showing the power of using these two approaches together. We categorize the literature based on proposed architectures, shared representations, and applications. Finally, we discuss several exciting future research directions and open problems.

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。

↑