检索结果-内蒙古大学图书馆

Dynamic spatial-temporal graph-driven machine remaining useful life prediction method using graph data augmentation

JOURNAL OF INTELLIGENT MANUFACTURING 2024年第1期35卷 355-366页

作者： Yang, Chaoying Liu, Jie Zhou, Kaibo Li, Xinyu Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat MOE Key Lab Image Informat Proc & Intelligent Con Wuhan 430074 Peoples R China Huazhong Univ Sci & Technol Sch Civil & Hydraul Engn Wuhan 430074 Peoples R China Huazhong Univ Sci & Technol State Key Lab Digital Mfg Equipment & Technol Wuhan 430074 Peoples R China

It is beneficial to maintain the normal operation of machines by conducting remaining useful life (RUL) prediction. Recently, graph data-driven machine RUL prediction methods have made a great success, since graph can model spatial and temporal dependencies of signals. However, the constructed graphs still have some limitations: (1) In the practical industrial production, the installation of multi-sensor networks is expensive and hard to achieve, so the single sensor is commonly used for data monitoring. However, most of these methods constructed graphs by establishing relationships between the different sensors, which are completely unsuitable for prediction tasks in single-sensor scenarios. (2) The quality of constructed graph is low, where the graph structure is fixed, failing in representing the machine degradation process. To overcome these limitations, a dynamic spatial-temporal (ST) graph-driven machine RUL prediction method using graph data augmentation (GDA) is proposed. The ST graph is constructed using short-time Fourier transform, capturing the frequency-domain and time-domain information hidden in the signals. Then, a GDA framework is designed to generate dynamic ST graphs, enlarging the structural differences of subgraphs. Subsequently, a GDA-based graph deep learning prediction model is constructed for dynamic ST graph-based RUL prediction, where an autoencoder-based graph embedding module is designed to replace simple Readout. Verification experiments are conducted on two case studies, and the results show that the proposed prediction method achieves a competitive performance.

关键词： Dynamic spatial-temporal graph Autoencoder Remaining useful life prediction graph data augmentation graph deep learning

来源：评论

学校读者我要写书评

暂无评论

graph Contrastive Learning with Constrained graph data augmentation

引用

NEURAL PROCESSING LETTERS 2023年第8期55卷 10705-10726页

作者： Xu, Shaowu Wang, Luo Jia, Xibin Beijing Univ Technol Fac Informat Technol Beijing 100124 Peoples R China

Studies on graph contrastive learning, which is an effective way of self-supervision, have achieved excellent experimental performance. Most existing methods generate two augmented views, and then perform feature learning on the two views through maximizing semantic consistency. Nevertheless, it is still challenging to generate optimal views to facilitate the graph construction that can reveal the essential association relations among nodes by graph contrastive learning. Considering that the extremely high mutual information between views is prone to have a negative effect on model training, a good choice is to add constraints to the graph data augmentation process. This paper proposes two constraint principles, low dissimilarity priority (LDP) and mutual exclusion (ME), to mitigate the mutual information between views and compress redundant parts of mutual information between views. LDP principle aims to reduce the mutual information between views at global scale, and ME principle works to reduce the mutual information at local scale. They are opposite and appropriate in different situations. Without loss of generality, the two proposed principles are applied to two well-performed graph contrastive methods, i.e. graphCL and GCA, and experimental results on 20 public benchmark datasets show that the models with the aid of the two proposed constraint principles achieve higher recognition accuracy.

关键词： Contrastive learning graph representation learning Self-supervised learning graph data augmentation augmentation principle

来源：评论

学校读者我要写书评

暂无评论

GMMDA: Gaussian mixture modeling of graph in latent space for graph data augmentation

引用

KNOWLEDGE AND INFORMATION SYSTEMS 2024年第12期66卷 7667-7695页

作者： Li, Yanjin Xu, Linchuan Yamanishi, Kenji Univ Tokyo Grad Sch Informat Sci & Technol Dept Math Informat 7-3-1 HongoBunkyo Ku Tokyo 1138656 Japan

graph data augmentation (GDA), which manipulates graph structure and/or attributes, has been demonstrated as an effective method for improving the generalization of graph neural networks on semi-supervised node classification. As a data augmentation technique, label preservation is critical, that is, node labels should not change after data manipulation. However, most existing methods overlook the label preservation requirements. Determining the label-preserving nature of a GDA method is highly challenging, owing to the non-Euclidean nature of the graph structure. In this study, for the first time, we formulate a label-preserving problem (LPP) in the context of GDA. The LPP is formulated as an optimization problem in which, given a fixed augmentation budget, the objective is to find an augmented graph with minimal difference in data distribution compared to the original graph. To solve the LPP problem, we propose GMMDA, a generative data augmentation (DA) method based on Gaussian mixture modeling (GMM) of a graph in a latent space. We designed a novel learning objective that jointly learns a low-dimensional graph representation and estimates the GMM. The learning is followed by sampling from the GMM, and the samples are converted back to the graph as additional nodes. To uphold label preservation, we designed a minimum description length (MDL)-based method to select a set of samples that produces the minimum shift in the data distribution captured by the GMM. Through experiments, we demonstrate that GMMDA can improve the performance of graph convolutional network on Cora, Citeseer and Pubmed by as much as 7.75%\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$7.75\%$$\end{document}, 8.75%\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepacka

关键词： graph data augmentation graph neural networks Gaussian mixture model Semi-supervised learning Minimum description length principle

来源：评论

学校读者我要写书评

暂无评论

Generalized heterophily graph data augmentation for node classification

引用

NEURAL NETWORKS 2023年 168卷 339-349页

作者： Tang, Bisheng Chen, Xiaojun Wang, Shaopu Xuan, Yuexin Zhao, Zhendong Univ Chinese Acad Sci Sch Cyber Secur Beijing 100190 Peoples R China Chinese Acad Sci Inst Informat Engn Shangdi St Shucun Rd 19 Beijing 100080 Peoples R China

graph data augmentations have demonstrated remarkable performance on homophilic graph neural networks (GNNs). Nevertheless, when transferred to a heterophilic graph, these augmentations are less effective for GNN models and lead to reduced performance. To address this issue, we propose a unified augmentation approach called GePHo, a regularization technique for heterophilic graph neural networks based on self-supervised learning, leveraging graph data augmentation to acquire extra information to guide model learning. Specifically, we propose to generate a pseudo-homophily graph that is type-agnostic, enabling us to apply GePHo to both homophilic and heterophilic graphs. Then, we regularize the neighbors with a sharpening technique for data augmentation and generate the auxiliary pseudo-labels to classify the original GNN's output, whose operations are to constrain the local and global node representation, respectively. Extensive experiments on three homophilic graph and six heterophilic graph datasets demonstrate the competitive effectiveness of GePHo in node classification task, and the ablation experiments verify the efficacy of our GePHo in graph data augmentation.(c) 2023 Elsevier Ltd. All rights reserved.

关键词： Heterophily graph data augmentation Self-supervised

来源：评论

学校读者我要写书评

暂无评论

GMMDA: Gaussian Mixture Modeling of graph in Latent Space for graph data augmentation 23

GMMDA: Gaussian Mixture Modeling of Graph in Latent Space fo...

引用

23rd IEEE International Conference on data Mining (IEEE ICDM)

作者： Li, Yanjin Xu, Linchuan Yamanishi, Kenji Univ Tokyo Grad Sch Informat Sci & Technol Dept Math Informat Tokyo Japan Hong Kong Polytech Univ Dept Comp Hong Kong Peoples R China

ISBN: (纸本)9798350307887

graph data augmentation (GDA), which manipulates graph structure and/or attributes, has been demonstrated as an effective method for improving the generalization of graph neural networks on semi -supervised node classification. As a data augmentation technique, label -preservation is critical, that is, node labels should not change after data manipulation. However, most existing methods overlook the label-preservation requirements. Determining the label-preserving nature of a GDA method is highly challenging, owing to the non-Euclidean nature of the graph structure. In this study, for the first time, we formulate a label-preserving problem (LPP) in the context of GDA. The LPP is formulated as an optimization problem in which, given a fixed augmentation budget, the objective is to find an augmented graph with minimal difference in data distribution compared to the original graph. To solve the LPP problem, we propose GMMDA, a generative data augmentation (DA) method based on Gaussian mixture modeling (GMM) of a graph in a latent space. The proposed GMMDA has three phases. First, a novel objective is designed to jointly learn a low-dimensional graph representation and estimate the GMM. The learning is followed by sampling from the GMM, and then the samples are converted hack to the graph as additional nodes. To uphold label preservation, we designed a minimum description length (MDL)-based method to select a set of samples that produces the minimum shift in the data distribution. Through experiments, we demonstrate that GMMDA can improve the performance of graph convolutional network on CORA, CITESEER and MIMED by as much as 7.75% 8.75% and 5.87%, respectively, significantly outperforming the state-of-the-art methods.

关键词： graph data augmentation graph neural networks Gaussian mixture model semi-supervised learning minimum description length principle

来源：评论

学校读者我要写书评

暂无评论

A Simple data augmentation for graph Classification: A Perspective of Equivariance and Invariance

引用

ACM TRANSACTIONS ON KNOWLEDGE DISCOVERY FROM data 2025年第2期19卷 1-24页

作者： Sui, Yongduo Wang, Shuyao Sun, Jie Liu, Zhiyuan Cui, Qing Li, Longfei Zhou, Jun Wang, Xiang He, Xiangnan Univ Sci & Technol China Hefei Peoples R China Natl Univ Singapore Singapore Singapore Ant Grp Hangzhou Peoples R China

In graph classification, the out-of-distribution (OOD) issue is attracting great attention. To address this issue, a prevailing idea is to learn stable features, on the assumption that they are substructures causally determining the label and that their relationship with the label is stable to the distributional uncertainty. In contrast, the complementary parts termed environmental features, fail to determine the label solely and hold varying relationships with the label, thus ascribed to the possible reason for the distribution shift. Existing generalization efforts mainly encourage the model's insensitivity to environmental features. While the sensitivity to stable features is promising to distinguish the crucial clues from the distributional uncertainty but largely unexplored. A paradigm of simultaneously exploring the sensitivity to stable features and insensitivity to environmental features is until-now lacking to achieve the generalizable graph classification, to the best of our knowledge. In this work, we conjecture that generalizable models should be sensitive to stable features and insensitive to environmental features. To this end, we propose a simple yet effective augmentation strategy for graph classification: Equivariant and Invariant Cross-data augmentation (EI-CDA). By employing equivariance, given a pair of input graphs, we first estimate their stable and environmental features via masks. Then, we linearly mix the estimated stable features of two graphs and encourage the model predictions faithfully reflect their mixed semantics. Meanwhile, by using invariance, we swap the estimated environmental features of two graphs and keep the predictions invariant. This simple yet effective strategy endows the models with both sensitivity to stable features and insensitivity to environmental features. Extensive experiments show that EI-CDA significantly improves performance and outperforms leading baselines. Our codes are available at: https://***/yongdu

关键词： graph data augmentation graph Classification Distribution Shift

来源：评论

学校读者我要写书评

暂无评论

LeDA-GNN: Learnable dual augmentation for graph neural networks

引用

EXPERT SYSTEMS WITH APPLICATIONS 2025年 268卷

作者： Liu, Gen Zhao, Zhongying Li, Chao Yu, Yanwei Shandong Univ Sci & Technol Coll Comp Sci & Engn Qingdao 266590 Shandong Peoples R China Shandong Univ Sci & Technol Coll Elect & Informat Engn Qingdao 266590 Shandong Peoples R China Ocean Univ China Coll Comp Sci & Technol Qingdao 266100 Shandong Peoples R China

graph Neural Networks (GNNs) have achieved remarkable success in various graph analyzing tasks. To handle the data sparsity and noise issues, data augmentation for GNNs is receiving a great deal of attention. However, the existing studies solely focus on either feature-level or topology-level augmentation. Moreover, these methods are mainly rule-based and exhibit limited generalization capabilities in cross-domain scenarios. To address these challenges, we propose a Learnable Dual augmentation method for graph Neural Networks (LeDA-GNN). Specifically, LeDA-GNN executes data augmentation on both node features and graph topology. Furthermore, it enables GNNs to encounter out-of-distribution data as much as possible, in a learnable manner, thus enhancing the generalization capability. In addition, the proposed LeDA-GNN is a plug-and-play method that can be easily integrated into any GNN-based model. The experimental results on the task of node classification fully demonstrate the effectiveness and superiority of this work. The source codes of this work are available at https://***/ZZY-graphMiningLab/LeDA-GNN.

关键词： graph representation learning graph neural networks graph data augmentation Node classification

来源：评论

学校读者我要写书评

暂无评论

Class-aware graph Siamese representation learning

引用

NEUROCOMPUTING 2025年 620卷

作者： Xu, Chengcheng Wang, Tianfeng Chen, Man Chen, Jun Pan, Zhisong Army Engn Univ PLA Command & Control Engn Coll Nanjing 210000 Jiangsu Peoples R China

Currently, two issues exist in the field of graph Siamese representation learning. First, the strategies for positive sample selection often impose strict constraints on the candidate set, which may result in critical class information being overlooked. Second, traditional data augmentation strategies often alter the topological structure of graphs, which may impact the representations of nodes. Therefore, we propose a class-aware graph Siamese representation learning method named CGSRL to address these issues. Inspired by the Jaccard coefficient, we design a node structural similarity calculation method that considers neighborhood information and average degree. In the selection of positive samples, we comprehensively considered both feature similarity and structural similarity, allowing for the selection of positive samples belonging to the same class over a wider range. Additionally, we craft a feature augmentation method that preserves the raw class information. Specifically, CGSRL enforces a discriminator to extract high-level semantics representing class information from the raw features through carefully designed consistency constraints. In turn, a generator synthesizes augmented features with the same class information as the raw features based on these high-level semantics. To validate the superiority of our proposed method, we conducted experiments in node classification and node clustering on five real-world graph datasets. The results consistently demonstrate that our proposed CGSRL method outperforms state-of-the-art baselines.

关键词： graph Siamese representation learning Positive sample selection graph data augmentation Class consistency

来源：评论

学校读者我要写书评

暂无评论

Asymmetric augmented paradigm-based graph neural architecture search

引用

INFORMATION PROCESSING & MANAGEMENT 2025年第1期62卷

作者： Wu, Zhenpeng Chen, Jiamin Al-Sabri, Raeed Oloulade, Babatounde Moctard Gao, Jianliang Cent South Univ Sch Comp Sci & Engn Changsha 410083 Hunan Peoples R China

In most scenarios of graph-based tasks, graph neural networks (GNNs) are trained end-to-end with labeled samples. Labeling graph samples, a time-consuming and expert-dependent process, leads to huge costs. graph data augmentations can provide a promising method to expand labeled samples cheaply. However, graph data augmentations will damage the capacity of GNNs to distinguish non-isomorphic graphs during the supervised graph representation learning process. How to utilize graph data augmentations to expand labeled samples while preserving the capacity of GNNs to distinguish non-isomorphic graphs is a challenging research problem. To address the above problem, we abstract a novel asymmetric augmented paradigm in this paper and theoretically prove that it offers a principled approach. The asymmetric augmented paradigm can preserve the capacity of GNNs to distinguish non-isomorphic graphs while utilizing augmented labeled samples to improve the generalization capacity of GNNs. To be specific, the asymmetric augmented paradigm will utilize similar yet distinct asymmetric weights to classify the real sample and augmented sample, respectively. To systemically explore the benefits of asymmetric augmented paradigm under different GNN architectures, rather than studying individual asymmetric augmented GNN (A(2)GNN) instance, we then develop an auto-search engine called Asymmetric Augmented graph Neural Architecture Search (A(2)GNAS) to save human efforts. We empirically validate our asymmetric augmented paradigm on multiple graph classification benchmarks, and demonstrate that representative A(2)GNN instances automatically discovered by our A(2)GNAS method achieve state-of-the-art performance compared with competitive baselines. Our codes are available at: https://***/csubigdata-Organization/A2GNAS.

关键词： graph neural network graph data augmentation Asymmetric augmented paradigm graph neural architecture search

来源：评论

学校读者我要写书评

暂无评论

Class-homophilic-based data augmentation for improving graph neural networks

引用

KNOWLEDGE-BASED SYSTEMS 2023年第1期269卷

作者： Duan, Rui Yan, Chungang Wang, Junli Jiang, Changjun Minist Educ Key Lab Embedded Syst & Serv Comp Shanghai 201804 Peoples R China Tongji Univ Natl Prov Minist Joint Collaborat Innovat Ctr Fin Shanghai 201804 Peoples R China Tongji Univ Shanghai 201804 Peoples R China

data augmentation has been shown to improve graph neural networks (GNNs). Existing graph data augmentation is achieved by adding or removing edges or changing the input node features due to graph data's complexity and non-Euclidean nature. However, the graphs generated by the above augmentation operations have similar or identical structures as the original graph, which leads to similar or identical structural information learned by GNNs from the original and augmented graphs. Two problems arise from this: restricted information and low applicability. To solve these problems, we propose Class-hOmophilic-based data augmentation (CODA), which improves existing GNNs by helping them learn adequate and extra structural class information, which is lacking in the original graph, and promotes GNNs' application to graphs with a large number of interclass edges. We first pretrain the GNNs and then design a new augmentation method that generates an approximate class-homophilic graph according to the pretrained GNNs. In the end, we design learnable node-level self-attention mechanisms with telescopic coefficients, which result in GNNs integrating the structural information of the two graphs in a more principled way to break the constraint of pretrained GNNs. Extensive experiments on various datasets show that augmentation via CODA improves performance and applicability across GNN architectures. The source code of CODA is publicly available at https: //***/graphNN/CODA.(c) 2023 Elsevier B.V. All rights reserved.

关键词： graph neural networks graph data augmentation Approximate class-homophilic graph Self-attention mechanisms

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：