文章A Survey of Adversarial Learning on Graphs提出了两种特殊的对抗训练策略,对所有节点进行保护的全局对抗训练Global-AT(global adversarial training)和保护目标节点免受攻击的目标标签对抗训练Target-AT(target label adversarial training)。在Global-AT中,首先选择目标节点对,更新 (t-1) 次对抗网络的邻接矩阵 \hat...
当然,貌似他们课题组的也对某些相关方面进行研究,有兴趣的可以参考综述《Graph Neural Networks: A Review of Methods and Applications》、《A comprehensive survey on graph neural networks》,将来我也会对这些文献进行总结分享。 文中还提到,Network embedding也是一个很火的研究点,可以实现将节点嵌入到低维的向量...
图自动编码器 (graph autoencoder,GAE),GAE 假设低秩图结构,并采用无监督方法进行节点表示学习。 图强化学习(graph reinforcement learning,Graph RL),图 RL 定义了基于图的动作和奖励,以在遵循约束的同时获得有关图任务的反馈。 图对抗方法(graph adversarial methods),图对抗方法采用对抗训练技术来增强基于图的模型...
论文标题:Data Augmentation for Deep Graph Learning: A Survey 论文作者:Kaize Ding, Zhe Xu, Hanghang Tong, Huan Liu 论文来源:2022, arXiv 论文地址:download 1 介绍 本文主要总结图数据增强,并对该领域的代表性方法做出归类分析。 DGL 存在的两个问题: ...
Adversarial examples are hot topics in the field of security in deep learning.The feature,generation methods,attack and defense methods of the adversarial examples are focuses of the current research on adversarial examples.This article explains the key technologies and theories of adversarial examples ...
A Comprehensive Survey of Graph-based Deep Learning Approaches for Anomaly Detection in Complex Distributed Systems Intelligent financial fraud detection practices in post-pandemic era A Comprehensive Survey on Graph Anomaly Detection with Deep Learning Anomaly Mining - Past, Present and Future Graph Comput...
Awesome Graph Causal Learning This repository contains a list of Graph Causal Learning resources. We also have a survey paper about Counterfactual Learning on Graphs. We include a Not for Graph section to introduce well-selected materials for beginners to learn causal-related concepts. We will try...
The rest of the paper is structured as follows: “Survey methodology” section describes the survey methodology, followed by “Types of learning” section which presents the state-of-the-art learning techniques. DL architectures are introduced in “Deep learning architectures” section, while “Lack...
卷积图神经网络(Convolutional graph neural networks, ConvGNNs)将网格数据的卷积操作推广到图数据。其主要思想是通过集合节点v自身的特征xv和相邻的特征xu来生成节点v的表示,其中u∈N(v)。与RecGNNs不同,ConvGNNs通过堆叠多个图卷积层来提取高级节点表示。卷积神经网络在构建其他复杂神经网络模型中发挥着核心作用。图...
【Arxiv-2021】【IEEE members/fellows】Graph Self-Supervised Learning: A Survey 核心要点 文章旨在对现有图神经网络的方法进行全面的总结和分类,并给出常用的数据集、评估基准、方法间的性能比较和开源代码链接。图的深度学习的热度与日俱增,但大多数工作集中在(半)监督学习上。对比标签的严重依赖导致模型泛化能力...