arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2403.01092cs.LG
收藏

成对对齐改进图域适应

Pairwise Alignment Improves Graph Domain Adaptation

Georgia Institute of Technology(佐治亚理工学院);University of Science and Technology of China(中国科学技术大学);University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校)

Shikun Liu, Deyu Zou, Han Zhao, Pan Li

AI总结:

针对图数据分布偏移问题,提出成对对齐方法,通过边权重缓解条件结构偏移和标签权重处理标签偏移,显著提升了图域适应在节点分类等任务中的性能。

摘要:

基于图的方法在许多现实世界应用中对互连对象的标签推断至关重要,但如果用于模型训练的图与用于测试的图存在显著差异,它们通常会面临泛化挑战。本研究深入探讨图域适应(GDA),以解决图数据上分布偏移的独特复杂性,其中互连的数据点在特征、标签,特别是连接模式上都会经历偏移。我们提出了一种新颖且具有理论原则的方法——成对对齐(Pair-Align),通过缓解条件结构偏移(CSS)和标签偏移(LS)来应对图结构偏移。Pair-Align 使用边权重重新校准相邻节点间的影响以处理 CSS,并使用标签权重调整分类损失以处理 LS。我们的方法在现实世界应用中展现出卓越的性能,包括社交网络中具有区域偏移的节点分类,以及粒子对撞实验中的堆积缓解任务。对于第一个应用,我们还为 GDA 研究整理了迄今为止最大的数据集。我们的方法在合成数据集和其他现有基准数据集上也表现出强劲的性能。

英文摘要:

Graph-based methods, pivotal for label inference over interconnected objects in many real-world applications, often encounter generalization challenges, if the graph used for model training differs significantly from the graph used for testing. This work delves into Graph Domain Adaptation (GDA) to address the unique complexities of distribution shifts over graph data, where interconnected data points experience shifts in features, labels, and in particular, connecting patterns. We propose a novel, theoretically principled method, Pairwise Alignment (Pair-Align) to counter graph structure shift by mitigating conditional structure shift (CSS) and label shift (LS). Pair-Align uses edge weights to recalibrate the influence among neighboring nodes to handle CSS and adjusts the classification loss with label weights to handle LS. Our method demonstrates superior performance in real-world applications, including node classification with region shift in social networks, and the pileup mitigation task in particle colliding experiments. For the first application, we also curate the largest dataset by far for GDA studies. Our method shows strong performance in synthetic and other existing benchmark 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。

↑