To address the above problems, a position-aware and subgraph enhanced dynamic graph contrastive learning method is proposed for discrete-time dynamic graphs. Firstly, the global snapshot is built based on the historical snapshots to express the stable pattern of the dynamic graph, and the random ...
Zhang G, Hu Z, Wen G, et al. Dynamic graph convolutional networks by semi-supervised contrastive learning[J]. Pattern Recognition, 2023, 139: 109486. 摘要导读 传统的图卷积网络(GCN)及其变体通常只通过数据集给出的拓扑结构传播节点信息。然而,给定的拓扑结构只能表示一定的关系,而忽略节点之间的一些相关...
Contrastive learning is a commonly used framework in the field of graph self-supervised learning, where models are trained by bringing positive samples closer together and pushing negative samples apart. Most existing graph contrastive learning models divide all nodes into positive and negative samples,...
2024 年 6 月 21 日,郑双佳课题组提出使用表型层面的扰动高内涵细胞显微镜图像,来辅助分子表征学习的跨模态学习方法。这种方法可以有效地在分子和表征之间架起桥梁,对药物研发具有重要意义。相关研究以「Cross-Modal Graph Contrastive Learning with Cellular Images」为题,发表在 Advanced Science 上。论文地址:https...
2024 年 6 月 21 日,郑双佳课题组提出使用表型层面的扰动高内涵细胞显微镜图像,来辅助分子表征学习的跨模态学习方法。这种方法可以有效地在分子和表征之间架起桥梁,对药物研发具有重要意义。相关研究以「Cross-Modal Graph Contrastive Learning with Cellular Images」为题,发表在 Advanced Science 上。
相关研究以「Cross-Modal Graph Contrastive Learning with Cellular Images」为题,发表在 Advanced Science 上。 2024 年 5 月 25 日,郑双佳课题组提出了多尺度学习框架 MUSE,有效地融合了原子结构和分子网络尺度之间的多尺度信息,展现了将计算药物发现扩展到其他尺度的潜力。相关研究以「A variational expectation-maxi...
2024 年 6 月 21 日,郑双佳课题组提出使用表型层面的扰动高内涵细胞显微镜图像,来辅助分子表征学习的跨模态学习方法。这种方法可以有效地在分子和表征之间架起桥梁,对药物研发具有重要意义。相关研究以「Cross-Modal Graph Contrastive Learning with Cellular Images」为题,发表在 Advanced Science 上。
Graph Embedding Dynamic Feature-based Supervised Contrastive Learning of Transient Stability for Changing Power Grid Topologies 准确的在线瞬间稳定预测对于在出现动乱时确保电源系统稳定性至关重要。 ht...
2024 年 6 月 21 日,郑双佳课题组提出使用表型层面的扰动高内涵细胞显微镜图像,来辅助分子表征学习的跨模态学习方法。这种方法可以有效地在分子和表征之间架起桥梁,对药物研发具有重要意义。相关研究以「Cross-Modal Graph Contrastive Learning with Cellular Images」为题,发表在 Advanced Science 上。
retrieved specific knowledge with general knowledge. Then contrastive learning is employed to improve visual and textual representations, which also promises the accuracy of our dynamic graph. Experiments on two popular benchmarks verify the effectiveness of our method in generating accurate and meaningful...