We study feature propagation on graph, an inference process involved in graph representation learning tasks. It's to spread the features over the whole graph to the $t$-th orders, thus to expand the end's features. The process has been successfully adopted in graph embedding or graph neural networks, however few works studied the convergence of feature propagation. Without convergence guarantees, it may lead to unexpected numerical overflows and task failures. In this paper, we first define the concept of feature propagation on graph formally, and then study its convergence conditions to equilibrium states. We further link feature propagation to several established approaches such as node2vec and structure2vec. In the end of this paper, we extend existing approaches from represent nodes to edges (edge2vec) and demonstrate its applications on fraud transaction detection in real world scenario. Experiments show that it is quite competitive.
Multi-perspective Feedback-attention Coupling Model for Continuous-time Dynamic Graphs
College of Electronic and Information Engineering, Tongji University（同济大学电子与信息工程学院）；School of Physics and Electronic-Electrical Engineering, Ningxia University（宁夏大学物理与电子电气工程学院）；Faculty of Anesthesiology, Naval Medical University（海军军医大学麻醉学院）
Randomness and signal propagation in physics-informed neural networks (PINNs): A neural PDE perspective
Department of Civil, Computer Science and Aeronautical Technologies Engineering（土木、计算机科学与航空技术工程系）；Engineering Mechanics Unit（力学单位）；Jawaharlal Nehru Centre for Advanced Scientific Research（贾瓦哈拉尔·奈尔中心高级科学研究中心）；Center for Life Nano-& Neuro-Science（生命纳米-神经科学中心）；Italian Institute of Technology (IIT)（意大利理工学院（IIT））
Department of Electrical Engineering, Tsinghua University（清华大学电子工程系）；Beijing National Research Center for Information Science and Technology（北京信息科学与技术国家研究中心）；State Key laboratory of Space Network and Communications（空间网络与通信国家重点实验室）
Accelerating Scalable Graph Neural Network Inference with Node-Adaptive Propagation
The University of Queensland（昆士兰大学）；Peking University（北京大学）；National Engineering Laboratory for Big Data Analysis and Applications（国家大数据分析与应用工程实验室）；Beijing University of Posts and Telecommunications（北京邮电大学）；Griffith University（格里菲斯大学）
Label Propagation on K-partite Graphs with Heterophily
Univ. of Southern California（南加州大学）；Fudan University（复旦大学）；Information Sciences Institute（信息科学研究所）
本文首次在异质性假设下研究了异质图中的标号传播问题。同构标签传播（即，两个连通节点共享相似标号）的问题已经得到了广泛的研究。不幸的是，现实生活中的网络是异构的，它们包含不同类型的顶点（例如，用户、图像、文本）和关系（例如，友谊、共同标记），并允许每个节点向其邻居传播相同和相反的标签副本。我们提出了一个$\mathcal{K}$ -partite标签传播模型来处理异构节点/关系和异构传播的神秘组合。在此基础上，我们提出了一种新的标签推理算法框架，该框架具有近线性时间复杂度的更新规则。由于真实网络随时间而变化，我们设计了一种增量方法，该方法支持新数据和证据的快速更新（例如，Ground truth labels），保证效率。我们还提供了一个效用函数，以自动确定是否是增量或重新建模的方法是有利的。在真实数据集上的大量实验验证了该方法的有效性和效率，以及其优于最先进的标签传播方法的优势。
Link prediction on multi-relational graphs from an influence propagation perspective
School of Information Science and Technology, Yunnan Normal University（云南师范大学信息科学与技术学院）；School of Computing, Macquarie University（麦考瑞大学计算学院）；School of Computer Science and Engineering, The University of New South Wales（新南威尔士大学计算机科学与工程学院）；School of Information Science and Engineering, Yunnan University（云南大学信息科学与工程学院）；School of Computer and Mathematical Sciences, Adelaide University（阿德莱德大学计算机与数学科学学院）
