arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2302.07491cs.LGcs.AIcs.SI
收藏

基于时间与结构强度对齐的自监督时序图学习

Self-Supervised Temporal Graph learning with Temporal and Structural Intensity Alignment

School of Computer, National University of Defense Technology(国防科技大学计算机学院);Medical Big Data Research Center, Chinese PLA General Hospital(中国人民解放军总医院医学大数据研究中心);College of Intelligence Science and Technology, National University of Defense Technology(国防科技大学智能科学与技术学院)

Meng Liu, Ke Liang, Yawei Zhao, Wenxuan Tu, Sihang Zhou, Xinbiao Gan, Xinwang Liu, Kunlun He

AI总结:

针对时序图学习方法忽略高阶结构信息导致性能次优的问题,提出自监督方法S2T,结合一阶时间与高阶结构信息计算条件强度并引入对齐损失,在多个数据集上最高提升10.13%性能。

摘要:

时序图学习旨在为包含动态信息的图任务生成高质量表示,近期引起了越来越多的关注。与静态图不同,时序图通常被组织为连续时间上的节点交互序列,而非邻接矩阵。大多数时序图学习方法通过结合历史邻域来建模当前交互。然而,此类方法仅考虑一阶时间信息,而忽略了关键的高阶结构信息,导致性能次优。为解决这一问题,我们提出了一种用于时序图学习的自监督方法S2T,该方法同时提取时间与结构信息以学习更具信息量的节点表示。值得注意的是,初始节点表示以不同方式结合一阶时间信息与高阶结构信息,以计算两种条件强度。随后引入对齐损失来优化节点表示,缩小两种强度之间的差距并使其更具信息量。具体而言,除了利用历史邻居序列建模时间信息外,我们还在局部和全局层面进一步考虑结构知识。在局部层面,通过聚合高阶邻居序列的特征来生成结构强度。在全局层面,基于所有节点生成全局表示,以根据不同节点上的活跃状态调整结构强度。大量实验表明,与最先进的竞争方法相比,所提出的模型S2T在多个数据集上实现了最高10.13%的性能提升。

英文摘要:

Temporal graph learning aims to generate high-quality representations for graph-based tasks with dynamic information, which has recently garnered increasing attention. In contrast to static graphs, temporal graphs are typically organized as node interaction sequences over continuous time rather than an adjacency matrix. Most temporal graph learning methods model current interactions by incorporating historical neighborhood. However, such methods only consider first-order temporal information while disregarding crucial high-order structural information, resulting in suboptimal performance. To address this issue, we propose a self-supervised method called S2T for temporal graph learning, which extracts both temporal and structural information to learn more informative node representations. Notably, the initial node representations combine first-order temporal and high-order structural information differently to calculate two conditional intensities. An alignment loss is then introduced to optimize the node representations, narrowing the gap between the two intensities and making them more informative. Concretely, in addition to modeling temporal information using historical neighbor sequences, we further consider structural knowledge at both local and global levels. At the local level, we generate structural intensity by aggregating features from high-order neighbor sequences. At the global level, a global representation is generated based on all nodes to adjust the structural intensity according to the active statuses on different nodes. Extensive experiments demonstrate that the proposed model S2T achieves at most 10.13% performance improvement compared with the state-of-the-art competitors on several datasets.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑