图自监督学习（Graph Self-supervised Learning）最新综述+Github代 …
在本综述中，我们扩展了最早出现在计算机视觉和自然语言处理领域的自监督学习，对现有的图自监督学习（Graph Self-supervised Learning，Graph SSL）技术进行了及时且全面的回顾。 具体地，本文 …
Self-Supervised Learning of Graph Neural Networks: A Unified Review
The self-supervision can contribute to the learning of graph encoders f by utilizing information from 𝒫 and minimizing a self-supervised loss ℒ s s l (f, 𝒫) determined by a specifically designed self-supervised …
GSSCL: A framework for Graph Self-Supervised Curriculum Learning …
Jan 1, 2025 · Abstract Graph self-supervised learning is an effective technique for learning common knowledge from unlabeled graph data through pretext tasks. To capture the interrelationships …
论文阅读 —— Graph Self-Supervised Learning: A Survey （自监督图 …
Apr 13, 2023 · 与 HTC 相似但不同的是，MICRO-Graph [101] 提出了一种不同但新颖的基于主题学习的采样作为隐式增强，以生成几个具有语义信息的子图，其中每个子图的嵌入更接近于整个图的表示。
Self-supervised graph contrastive learning with diffusion …
Apr 1, 2025 · A self-supervised graph contrastive learning framework with diffusion augmentation is developed for fMRI analysis and brain disease detection, which effectively tackles the small-sample …
Linear projection fused graph-based semi-supervised learning on …
Jul 12, 2025 · In recent years, the surge in data-driven applications across various domains has spurred heightened interest in semi-supervised learning applied to graphs. This surge is attributed to the …
Semi-supervised learning-based virtual adversarial training on graph ...
Mar 1, 2025 · The key idea behind GVAT is that we focus on a semi-supervised learning-based virtual adversarial training on molecular graph structure, which incorporates the information in both labeled …
Self-supervised Graph-level Representation Learning with Adversarial ...
Nov 14, 2023 · We optimize the parameters of the graph encoding branch and adversarial generation branch alternately. Extensive experiments on 14 real-world benchmarks on both graph classification …
Graph Self-Supervised Learning: A Survey | IEEE Transactions on ...
Jun 1, 2023 · To address these issues, self-supervised learning (SSL), which extracts informative knowledge through well-designed pretext tasks without relying on manual labels, has become a …
Semi-supervised graph contrastive learning for emotion recognition ...
Dec 1, 2025 · To address these issues, we propose a semi-supervised graph contrastive learning (SGCL) model for EEG-based emotion recognition, leveraging the abundance of unlabeled EEG data …