arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2212.02810cs.LG
收藏

基于迭代图重构的数据插补

Data Imputation with Iterative Graph Reconstruction

Central South University(中南大学)

Jiajun Zhong, Weiwei Ye, Ning Gui

AI总结:

针对现有图神经网络数据插补方法未区分样本关联的问题,提出带朋友网络迭代重构的IGRM框架,通过差异化消息传递优化插补,在8个基准数据集上误差显著优于9种基线方法。

摘要:

有效的数据插补需要从“普通”表格数据中挖掘丰富潜在“结构”的能力。近期基于图神经网络的数据插补方案通过将表格数据直接转化为二部图,展现出强大的结构学习潜力。然而,由于缺乏样本间的关联,这些方案对所有样本一视同仁,这与“相似样本应能提供更多缺失值相关信息”的重要观察相悖。\n本文提出一种用于缺失数据插补的新型迭代图生成与重构框架IGRM(Iterative graph Generation and Reconstruction framework for Missing data imputation)。我们并未对所有样本同等对待,而是引入“朋友网络”的概念来表征样本间的不同关联。为在存在缺失数据的情况下生成准确的朋友网络,我们设计了一种端到端的朋友网络重构方案,可在插补学习过程中持续优化朋友网络。优化后的朋友网络表征反过来又通过差异化消息传递进一步优化数据插补过程。\n在8个基准数据集上的实验结果表明,与9种基线方法相比,IGRM的平均绝对误差降低了39.13%,较次优方法降低9.04%。我们的代码可在https://github.com/G-AILab/IGRM获取。

英文摘要:

Effective data imputation demands rich latent ``structure" discovery capabilities from ``plain" tabular data. Recent advances in graph neural networks-based data imputation solutions show their strong structure learning potential by directly translating tabular data as bipartite graphs. However, due to a lack of relations between samples, those solutions treat all samples equally which is against one important observation: ``similar sample should give more information about missing values." This paper presents a novel Iterative graph Generation and Reconstruction framework for Missing data imputation(IGRM). Instead of treating all samples equally, we introduce the concept: ``friend networks" to represent different relations among samples. To generate an accurate friend network with missing data, an end-to-end friend network reconstruction solution is designed to allow for continuous friend network optimization during imputation learning. The representation of the optimized friend network, in turn, is used to further optimize the data imputation process with differentiated message passing. Experiment results on eight benchmark datasets show that IGRM yields 39.13% lower mean absolute error compared with nine baselines and 9.04% lower than the second-best. Our code is available at https://github.com/G-AILab/IGRM.

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。

↑