检索结果-内蒙古大学图书馆

Interdependence-Adaptive Mutual Information Maximization for graph contrastive learning

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING 2024年第12期36卷 8556-8567页

作者： Sun, Qingqiang Wang, Kai Zhang, Wenjie Cheng, Peng Lin, Xuemin Great Bay Univ Sch Engn Dongguan 523000 Peoples R China Cent South Univ Sch Automat Changsha 410083 Peoples R China Univ New South Wales Sch Comp Sci & Engn Sydney NSW 2052 Australia East China Normal Univ Sch Software Engn Shanghai 200062 Peoples R China Shanghai Jiao Tong Univ Univ Antai Coll Econ & Management Shanghai 200030 Peoples R China

Despite remarkable advancements in graph contrastive learning techniques, the identification of interdependent relationships when maximizing cross-view mutual information remains a challenging issue, primarily due to the complexity of graph topology. In this study, we propose to formulate cross-view interdependence from the innovative perspective of information flow. Accordingly, IDEAL, a simple yet effective framework, is proposed for interdependence-adaptive graph contrastive learning. Compared with existing methods, IDEAL concurrently addresses same-node and distinct-node interdependence, circumvents the reliance on additional distribution mining techniques, and is augmentation-aware. Besides, the objective of IDEAL takes advantage of both contrastive and generative learning objectives and is thus capable of learning a uniform embedding distribution while retaining essential semantic information. The effectiveness of IDEAL is validated by extensive empirical evidence. It consistently outperforms state-of-the-art self-supervised methods by considerable margins across seven benchmark datasets with diverse scales and properties and, at the same time, showcases promising training efficiency.

关键词： contrastive learning Semantics Task analysis Mutual information Topology Representation learning Training Cross-view interdependence graph contrastive learning negative mining node representation learning self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Seeking False Hard Negatives for graph contrastive learning

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 2024年第8期34卷 7454-7466页

作者： Liu, Xin Qian, Biao Liu, Haipeng Guo, Dan Wang, Yang Wang, Meng Hefei Univ Technol Sch Comp Sci & Informat Engn Hefei 230601 Peoples R China

graph contrastive learning (GCL) has achieved great success in self-supervised representation learning throughout positive and negative pairs based on graph neural networks (GNNs), where one critical issue lies in how to handle the false hard negatives that share the large similarity to the same referenced class as the anchor, which is critical to message passing of GNNs to exploit the graph structure. However, the existing arts either mistakenly identify or miss the false hard negatives, hence resulting into poor node representation. Building on this, there are several crucial bottlenecks - Where do false hard negatives exist upon the anchor? How to well seek false hard negatives? Whether are more false hard negatives better? To answer these questions, in this paper, we propose a novel Locally Weighted graph contrastive learning method, named LocWGCL, while revealing that false hard negatives are primarily distributed in the first-order and second-order neighborhoods of the anchor. Benefiting from the tightness between the first-order nodes and the anchor, representation similarity is calculated to select false hard negatives. For the second-order case, false hard negatives are identified, such that they share the similar passed message with the anchor over the common first-order nodes, along with the large similarity. Upon the seeking process, we devise a weighted strategy to false hard negatives for better node representation. Empirical studies verify the advantages of LocWGCL over the state-of-the-arts on six benchmarks.

关键词： graph contrastive learning graph neural networks false hard negatives message passing graph contrastive learning graph neural networks false hard negatives message passing

来源：评论

学校读者我要写书评

暂无评论

Breaking the curse of dimensional collapse in graph contrastive learning: A whitening perspective

引用

INFORMATION SCIENCES 2024年 657卷

作者： Tao, Yang Guo, Kai Zheng, Yizhen Pan, Shirui Cao, Xiaofeng Chang, Yi Jilin Univ Sch Artificial Intelligence Changchun 130012 Jilin Province Peoples R China Monash Univ Dept Data Sci & AI Melbourne Vic Australia Griffith Univ Sch Informat & Commun Technol Brisbane Qld 4222 Australia

Dimensional collapse in graph contrastive learning (GCL) confines node embeddings to their lower-dimensional subspace, diminishing their distinguishability. However, the causes and solutions of this curse remain relatively underexplored. In statistics, whitening presents a powerful tool to eliminate correlations among multiple variables. This motivates us to relieve the dimensional collapse of GCL from a whitening perspective. In this paper, we propose an intuitive analysis suggesting that high similarity scores of node embeddings may cause dimensional collapse, providing more evidence for its presence. Considering the success of whitening in statistics, we introduce a new plug-and-play module called the Whitening graph contrastive learning (WGCL) to address the dimensional collapse issue in existing GCL methods. WGCL plugin standardises the covariance matrices of dimensions, eliminating correlations among node embeddings' dimensions. Additionally, we enhance the conventional GCL training objective by introducing a mutual information maximisation loss between input features and node embeddings to maintain information capacity. Our experiments demonstrate that WGCL effectively addresses dimensional collapse, leading to an average improvement of 0.93% (up to 2.0%) in classification accuracy across three GCL backbones on nine widely-used datasets. The code to reproduce the experiments is available at https://github .com /acboyty /WGCL.

关键词： graph contrastive learning Dimensional collapse Whitening

来源：评论

学校读者我要写书评

暂无评论

An Efficient and Lightweight Spectral-Spatial Feature graph contrastive learning Framework for Hyperspectral Image Clustering

引用

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 2024年 62卷

作者： Yang, Aitao Li, Min Ding, Yao Xiao, Xiongwu He, Yujie Xian Res Inst High Technol Xian 710025 Peoples R China Wuhan Univ State Key Lab Informat Engn Surveying Mapping & Re Wuhan 430079 Peoples R China

Due to the scarcity of prior information and the high complexity of spectral data, hyperspectral image (HSI) clustering presents a significant challenge. Although recent deep clustering methods have demonstrated remarkable performance, their intricate network structures and poor robustness hinder their practical application. To address this issue, we propose an efficient and lightweight spectral-spatial feature graph contrastive learning (S2GCL) framework for robust HSI clustering. Specifically, we have designed a novel spectral-spatial feature encoder that fully leverages the information in HSI by incorporating both spatial structure and spectral similarity matrices. To establish a lightweight model, we implement several effective designs: First, S2GCL eliminates the commonly used data augmentation and discriminator in GCL during the generation of positive embeddings. Second, we use a multilayer perceptron (MLP) to produce low-dimensional embeddings instead of relying on graph convolutional networks (GCNs). Third, negative embeddings are generated through row-shuffling, avoiding the use of neural networks. Finally, we propose a multiple boundary loss function to extract complementary information from spatial structures and neighboring nodes, while also constraining the interclass differences between positive and negative examples. We conducted extensive experiments on four publicly available datasets and compared S2GCL with state-of-the-art clustering methods. The results indicate that S2GCL achieves satisfactory performance. The code for S2GCL will be released at https://***/ahappyyang/S2GCL.

关键词： Feature extraction contrastive learning Land surface Semantics Clustering methods Data mining Complexity theory Training Hyperspectral imaging graph convolutional networks graph contrastive learning graph neural network hyperspectral image (HSI) clustering lightweight

来源：评论

学校读者我要写书评

暂无评论

Multi-Network graph contrastive learning for Cancer Driver Gene Identification

引用

IEEE TRANSACTIONS ON NETWORK SCIENCE AND ENGINEERING 2024年第4期11卷 3430-3440页

作者： Peng, Wei Zhou, Zhengnan Dai, Wei Yu, Ning Wang, Jianxin Kunming Univ Sci & Technol Fac Informat Engn & Automat Kunming 650093 Peoples R China Kunming Univ Sci & Technol Comp Technol Applicat Key Lab Yunnan Prov Kunming 650093 Peoples R China State Univ New York Dept Comp Sci Coll Brockport Brockport NY 14422 USA Cent South Univ Sch Comp Sci & Engn Changsha 410083 Peoples R China Cent South Univ Hunan Prov Key Lab Bioinformat Changsha 410083 Peoples R China

Identifying driver genes contributing to the occurrence and development of cancers plays a critical role in cancer research and treatment. Some recent computational approaches identify cancer-driver genes based on gene networks, assuming that cancer-driver genes perform essential functions in gene networks. Due to the noise in gene function networks, many works focus on integrating gene networks derived from multi-omics datasets to improve the accuracy of cancer driver gene detection. However, most of them ignore the information interactions between these multi-omics datasets. In this work, we propose MNGCL, a Multi-Network graph contrastive learning method to identify cancer driver genes. It first constructs three gene networks as different views based on protein interactions, gene semantic similarities, and gene co-occurrence in signaling pathways. Then, we perform data augmentation of these gene networks and input them into a graph contrastive learning (GCL) encoder with shared parameters to learn consistent gene feature representation in different networks from a holistic perspective. After that, the gene features from the GCL encoder are passed through three different graph convolutional networks to generate the unique gene feature representations in the three networks. Finally, we used a logistic regression model to fuse the gene feature representations generated in each network to predict cancer driver genes. The experimental results show that MNGCL improves the area under the ROC curve (AUC) and the area under the precision-recall curve (AUPRC) to a greater extent than the existing methods in identifying driver genes for both pan-cancer and single-type cancers. Furthermore, the ablation studies show that our model capturing dependencies and interactions between gene networks provided a more comprehensive perspective on the molecular mechanisms underlying cancer and improved the accuracy of cancer driver identification.

关键词： Cancer Self-supervised learning Semantics Bioinformatics Proteins Data augmentation Predictive models Cancer driver genes multi-view gene network graph contrastive learning network integration

来源：评论

学校读者我要写书评

暂无评论

Federal graph contrastive learning With Secure Cross-Device Validation

引用

IEEE TRANSACTIONS ON MOBILE COMPUTING 2024年第12期23卷 14145-14158页

作者： Wang, Tingqi Zheng, Xu Zhang, Jinchuan Tian, Ling Univ Elect Sci & Technol China Sch Comp Sci & Engn Chengdu 611731 Sichuan Peoples R China

Distributed mobile devices collect unlabeled graph data from environment. Introducing popular graph contrastive learning (GCL) methods can learn node representations better. However, training high-performance GCL requires large-scale data and graph data collected by the single device is insufficient. Meanwhile, transmitting local data for centralized training suffers from non-negotiate privacy leakage and bandwidth consumption. Federated learning (FL), as a distributed learning paradigm, is commonly used for such issues. Nevertheless, direct combination of FL and GCL struggles to supplement global graph information. This absence results in neighbor information missing, thus causing the local GCL to learn biased node representations. Moreover, the combination also triggers potential gradient explosion owing to the lack of unified learning criteria. In this paper, we propose a federal GCL framework that complements missing structural information and provides unified learning criteria. The key idea is to achieve cross-client node alignment on server through local graph structural importance to reason about the global graph information. We design a hierarchical structural importance scoring method to comprehensively evaluate structural importance, thus server performs effective cross-client aggregation while maintaining local graph privacy. We demonstrate the security and prove the bandwidth-reducing advantage of the proposed framework. Extensive experiments on 3 datasets show the superior performance of our method.

关键词： Training Servers Mobile handsets Distributed databases Computational modeling graph neural networks Federated learning Distributed knowledge base federated learning graph contrastive learning

来源：评论

学校读者我要写书评

暂无评论

Multi-aspect graph contrastive learning for Review-enhanced Recommendation

引用

ACM TRANSACTIONS ON INFORMATION SYSTEMS 2024年第2期42卷 1-29页

作者： Wang, Ke Zhu, Yanmin Zang, Tianzi Wang, Chunyang Liu, Kuan Ma, Peibo Shanghai Jiao Tong Univ Dept Comp Sci & Engn 800 Dongchuan Rd Shanghai 200240 Peoples R China

Review-based recommender systems explore semantic aspects of users' preferences by incorporating user-generated reviews into rating-based models. Recent works have demonstrated the potential of review information to improve the recommendation capacity. However, most existing studies rely on optimizing review-based representation learning part, thus failing to explicitly capture the fine-grained semantic aspects, and also ignoring the intrinsic correlation between ratings and reviews. To address these problems, we propose a multi-aspect graph contrastive learning framework, namedMAGCL, with three distinctive designs: (i) a multi-aspect representation learning module, which projects semantic relations to different subspaces by decoupling review information, and then obtains high-order decoupled representations in each aspect via graph encoder. (ii) the contrastive learning module performs graph contrastive learning to capture the correlation between rating and review patterns, which utilize unlabeled data to generate self-supervised signals and, in turn, relieve the data sparsity problem of supervision signals. (iii) the multi-task learning module conducts joint training to learn high-order structure-aware yet self-discriminative node representations by combining recommendation task and self-supervised task, which helps alleviate the over-smoothing problem. Extensive experiments are conducted on four real-world review datasets and the results show the superiority of the proposed framework MAGCL compared with several state of the arts. We also provide further analysis on multi-aspect representations and graph contrastive learning to verify the advantage of proposed framework.

关键词： Multi-aspect representations self-supervised signals graph contrastive learning multi-task learning

来源：评论

学校读者我要写书评

暂无评论

Adaptive graph contrastive learning with joint optimization of data augmentation and graph encoder

引用

KNOWLEDGE AND INFORMATION SYSTEMS 2024年第3期66卷 1657-1681页

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

graph contrastive learning (GCL) has been successfully used to solve the problem of the huge cost of graph data annotation, such as labor cost, time cost, and professional knowledge cost. Recent works have focused on improving the generalization performance of GCL with automated data augmentation. However, GCL methods with automated data augmentation encode the graph representation using fixed graph encoders, which will result in performance loss. To overcome this limitation, we propose Adaptive graph contrastive learning with joint optimization of data Augmentation and graph Encoder (AdaAE). AdaAE is the first method to learn to adapt the graph encoder for each dataset for GCL with automated data augmentation. Specifically, we design a unified GCL search space, which means that we treat both data augmentation and graph encoder as architecture components. AdaAE employs the adaptive architecture optimizer based on a differentiable search method to learn the sample probability distribution of each architecture component. The adaptive architecture optimizer generates an operation of each architecture component based on the sample probability distribution to construct the GCL model. Then, the validation results of the GCL model are used as the feedback signal of the adaptive architecture optimizer to optimize the sample probability distribution of each architecture component. Extensive experiments demonstrate that AdaAE performs more superior than the state-of-the-art baselines. Furthermore, the visualization results confirm that AdaAE can clearly distinguish different classes in the projection space.

关键词： graph contrastive learning graph neural network graph neural architecture search Data augmentation

来源：评论

学校读者我要写书评

暂无评论

Accurate Spatial Heterogeneity Dissection and Gene Regulation Interpretation for Spatial Transcriptomics using Dual graph contrastive learning

引用

ADVANCED SCIENCE 2024年第3期12卷 e2410081页

作者： Yu, Zhuohan Yang, Yuning Chen, Xingjian Wong, Ka-Chun Zhang, Zhaolei Zhao, Yuming Li, Xiangtao Jilin Univ Sch Artificial Intelligence Jilin 130012 Peoples R China Univ Toronto Terrence Donnelly Ctr Cellular & Biomol Res Toronto ON M5S 3E1 Canada Massachusetts Gen Hosp Cutaneous Biol Res Ctr Harvard Med Sch Boston MA 02115 USA City Univ Hong Kong Dept Comp Sci Hong Kong 999077 Peoples R China Northeast Forestry Univ Coll Comp & Control Engn Harbin 150040 Peoples R China

Recent advances in spatial transcriptomics have enabled simultaneous preservation of high-throughput gene expression profiles and the spatial context, enabling high-resolution exploration of distinct regional characterization in tissue. To effectively understand the underlying biological mechanisms within tissue microenvironments, there is a requisite for methods that can accurately capture external spatial heterogeneity and interpret internal gene regulation from spatial transcriptomics data. However, current methods for region identification often lack the simultaneous characterizing of spatial structure and gene regulation, thereby limiting the ability of spatial dissection and gene interpretation. Here, stDCL is developed, a dual graph contrastive learning method to identify spatial domains and interpret gene regulation in spatial transcriptomics data. stDCL adaptively incorporates gene expression data and spatial information via a graph embedding autoencoder, thereby preserving critical information within the latent embedding representations. In addition, dual graph contrastive learning is proposed to train the model, ensuring that the latent embedding representation closely resembles the actual spatial distribution and exhibits cluster similarity. Benchmarking stDCL against other state-of-the-art clustering methods using complex cortex datasets demonstrates its superior accuracy and effectiveness in identifying spatial domains. Our analysis of the imputation matrices generated by stDCL reveals its capability to reconstruct spatial hierarchical structures and refine differential expression assessment. Furthermore, it is demonstrated that the versatility of stDCL in interpretability of gene regulation, spatial heterogeneity at high resolution, and embryonic developmental patterns. In addition, it is also showed that stDCL can successfully annotate disease-associated astrocyte subtypes in Alzheimer's disease and unravel multiple relevant pathways and regulatory m

关键词： dual graph contrastive learning gene regulation graph contrastive learning spatial heterogeneity

来源：评论

学校读者我要写书评

暂无评论

TP-GCL: graph contrastive learning from the tensor perspective

引用

FRONTIERS IN NEUROROBOTICS 2024年 18卷 1381084页

作者： Li, Mingyuan Meng, Lei Ye, Zhonglin Yang, Yanglin Cao, Shujuan Xiao, Yuzhi Zhao, Haixing Qinghai Normal Univ Coll Comp Xining Peoples R China State Key Lab Tibetan Intelligent Informat Proc & Xining Peoples R China

graph Neural Networks (GNNs) have demonstrated significant potential as powerful tools for handling graph data in various fields. However, traditional GNNs often encounter limitations in information capture and generalization when dealing with complex and high-order graph structures. Concurrently, the sparse labeling phenomenon in graph data poses challenges in practical applications. To address these issues, we propose a novel graph contrastive learning method, TP-GCL, based on a tensor perspective. The objective is to overcome the limitations of traditional GNNs in modeling complex structures and addressing the issue of sparse labels. Firstly, we transform ordinary graphs into hypergraphs through clique expansion and employ high-order adjacency tensors to represent hypergraphs, aiming to comprehensively capture their complex structural information. Secondly, we introduce a contrastive learning framework, using the original graph as the anchor, to further explore the differences and similarities between the anchor graph and the tensorized hypergraph. This process effectively extracts crucial structural features from graph data. Experimental results demonstrate that TP-GCL achieves significant performance improvements compared to baseline methods across multiple public datasets, particularly showcasing enhanced generalization capabilities and effectiveness in handling complex graph structures and sparse labeled data.

关键词： graph neural network graph contrastive learning complex structure hypergraph high-order adjacency tensor

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：