需要注意的是GNNs学习图的表示和CNNs学习图片的表示非常不同,因为GNNs在graph universe和representation space之间是非单射的,GNNs的表达能力被1-WL test限制。所以我们需要基于1-WL test给出的等价关系定义一个graph universe的商空间(quotient space)。 ---手动分割线 --- 这里先给出一些定义的解释: 商空间(Quoti...
其中 Graph representation learning 包括了 DeepWalk、LINE、node2vec、LightGCN;Graph contrastive learning 中包括了 SimpleGCL、DGI、GraphCL、GRACE、SGL;Graph generative and adversarial learning 中选择了 GraphGAN、AD-GCL、GraphMAE。 5.1.3 Parameter Settings. 5.1.4 Metrics. 5.2 Node Classification (RQ1) ...
Self-supervised graph representation learning has recently shown considerable promise in a range of fields, including bioinformatics and social networks. A large number of graph contrastive learning approaches have shown promising performance for representation learning on graphs, which train models by ...
To improve the robustness of graph representation learning, we propose a novel Graph Adversarial Contrastive Learning framework (GraphACL) by introducing adversarial augmentations into graph self-supervised learning. In this framework, we maximize the mutual information between local and global ...
[NeurIPS 2021] ADGCL : Adversarial Graph Augmentation to Improve Graph Contrastive Learning | Paper | Talk | Poster Introduction This repo contains the Pytorch [1] implementation of Adversarial Graph Contrastive Learning (AD-GCL) principle instantiated with learnable edge dropping augmentation. The paper...
Paper tables with annotated results for On the Adversarial Robustness of Graph Contrastive Learning Methods
The recently developed unsupervised graph representation learning approaches apply contrastive learninginto graph-structured data and achieve promising performance. However, these methods mainly focus ongraph augmentation for positive samples, while the negative mining strategies for graph contrastive learningare ...
This is the implementation of Spatial-Temporal Graph Learning with Adversarial Contrastive Adaptation (ICML'23) in the following paper:RequirementsPytorch = 1.7.0 and Tensorflow = 1.15.3 (crime prediction task (ST-SHN))DataMoblity data, crime data, census block data, POI data, house price of ...
* Contrastive learning-based computational histopathology predict differential expression of cancer driver genes* 链接: arxiv.org/abs/2204.1199* 作者: Haojue Huang,Gongming Zhou,Xuejun Liu,Lei Deng,Chen Wu,Dachuan Zhang,Hui Liu* 摘要: 数字病理分析作为用于癌症诊断的主要检查。最近,从病理图像中进行深度...
(30) 基于图结构的鲁棒性多标签特征选择方法,考虑模糊依赖性和特征相互作用A Robust Multilabel Feature Selection Approach Based on Graph Structure Considering Fuzzy Dependency and Feature Interaction (31) 基于事件触发的有限时间自适应模糊控制,适用于状态无法测量的随机非线性系统Event-Trigger-Based Finite-Time ...