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 2202.10679cs.LGcs.AI
收藏

基于物理的图学习

Physics-Informed Graph Learning

Federation University Australia(澳大利亚联邦大学);Arizona State University(亚利桑那州立大学)

Ciyuan Peng, Feng Xia, Vidya Saikrishna, Huan Liu

摘要:

近年来,图学习得到了迅速的发展,在许多不同的领域都有着广泛的应用。相关的主要挑战是图形数据的数量和复杂性。图学习模型无法有效地学习图信息。为了弥补这种无效性,物理知情图学习(PIGL)正在出现。PIGL在执行图形学习时结合了物理规则,这具有巨大的好处。本文对PIGL方法进行了系统的综述。我们首先介绍了一个统一的框架图学习模型,然后检查现有的PIGL方法的统一框架。我们还讨论了PIGL未来的几个挑战。这份调查报告预计将促进与PIGL有关的创新研究和开发活动。

英文摘要:

An expeditious development of graph learning in recent years has found innumerable applications in several diversified fields. Of the main associated challenges are the volume and complexity of graph data. The graph learning models suffer from the inability to efficiently learn graph information. In order to indemnify this inefficacy, physics-informed graph learning (PIGL) is emerging. PIGL incorporates physics rules while performing graph learning, which has enormous benefits. This paper presents a systematic review of PIGL methods. We begin with introducing a unified framework of graph learning models followed by examining existing PIGL methods in relation to the unified framework. We also discuss several future challenges for PIGL. This survey paper is expected to stimulate innovative research and development activities pertaining to PIGL.

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。

↑