A dynamic attention mechanism is employed in the Graph Attention Networks (GAT) for modeling the KG. Combined with the nonlinear transformation and Momentum Contrast (Moco) strategy for contrastive learning, it
论文标题:Multi-Level Graph Contrastive Learning论文作者:Pengpeng Shao, Tong Liu, Dawei Zhang, J. Tao, Feihu Che, Guohua Yang论文来源:2021, Neurocomputing论文地址:download论文代码:download 1 Introduction本文贡献:提出多层次图对比学习框架:联合节点级和图级对比学习; 引入KNN 图提取语义信息;...
a multi-source medical knowledge augmented medication prediction network, which incorporates a multi-level graph contrastive learning framework. To the best of our knowledge, our model is the first to capture valuable relations between medical codes and augment their representations using a cascaded...
Feature-level Contrastive Learning 论文发现直接探索用户-项目图无法全面捕捉用户间(项目间)的共同作用信息。为此,我们构建了用户-用户(项目-项目)图,以便有效捕捉用户(项目)间的共同作用信号。特征学习通过对比机制捕捉来自用户-项目视图和用户-用户视图的信息并相互补充,从而实现自监督信号学习。最后,从提取的用户特征...
X(l)=GraphEncoder(X,A)=D−12AD−12X(l−1) 3. 方法 如图2 所示为方法框架图。它包括四个主要部分: 1)图构建层。它通过聚合用户行为序列来构造用户-商品、用户-用户和商品-商品图; 2)兴趣级别的对比学习层。它首先从用户行为序列中学习用户当前的兴趣,从用户-商品图中学习用户的总体兴趣,然后...
Further, the Geometric Multi-Scale Pixel-level Contrastive Learning (GMPCL) approach, which enhances the geometric representation of features is proposed using GMPCL loss and separates the geometric representations of foreground and background features of objects at the pixel level. The performance ...
This paper proposes a generic unsupervised Skeleton Prototype Contrastive learning paradigm with Multi-level Graph Relation learning (SPC-MGR) to learn effective representations from unlabeled skeletons to perform person re-ID. Specifically, we first construct unified multi-level skeleton graphs to fully ...
contrastive learning network, dubbed CoCoNet, to realize infrared and visible image fusion in an end-to-end manner. Concretely, to simultaneously retain typical features from both modalities and to avoid artifacts emerging on the fused result, we develop a coupled contrastive constraint in our loss ...
DCMSL: Dual influenced community strength-boosted multi-scale graph contrastive learning 2024, Knowledge-Based Systems Citation Excerpt : Graph Neural Networks (GNNs) are designed to process non-Euclidean data, serving as a powerful tool applicable across various domains, including recommendation systems ...
Multi-task learning MSC 41A05 41A10 65D05 65D17 1. Introduction The novelcoronavirusdisease 2019 (COVID-19) caused bysevere acute respiratory syndrome coronavirus2 (SARS-CoV-2) has emerged as one of the deadliest viruses of the century, resulting in about 137 million people infected with over...