arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2305.16618cs.LGcs.SI
收藏

基于置信度的部分已知特征图的特征填补

Confidence-Based Feature Imputation for Graphs with Partially Known Features

Seoul National University(首尔国立大学)

Daeho Um, Jiwoong Park, Seulki Park, Jin Young Choi

AI总结:

本文针对图学习中的高缺失率特征填补问题,提出基于逐通道伪置信度的填补方案,通过节点间扩散和通道间传播,在99.5%缺失率下仍取得最优节点分类和链路预测性能。

摘要:

本文研究了图学习任务中的缺失特征填补问题。此前已有多种方法处理带有缺失特征的图学习任务,但在特征缺失率较高的情况下,它们无法避免显著的性能下降。为克服这一局限,我们引入了一种新的节点特征逐通道置信度概念,将其分配给节点每个被填补的通道特征,以反映填补的确定性。随后,我们利用缺失特征节点与其最近已知特征节点之间的逐通道最短路径距离来设计伪置信度,以在实际学习过程中替代不可用的真实置信度。基于该伪置信度,我们提出了一种新的特征填补方案,该方案执行逐通道的节点间扩散和逐节点的通道间传播。该方案即使在极高缺失率(如99.5%)下也能保持稳定,并在多个包含高缺失率特征的数据集上,在半监督节点分类和链路预测任务中均达到了最先进的准确率。代码可在 https://github.com/daehoum1/pcfi 获取。

英文摘要:

This paper investigates a missing feature imputation problem for graph learning tasks. Several methods have previously addressed learning tasks on graphs with missing features. However, in cases of high rates of missing features, they were unable to avoid significant performance degradation. To overcome this limitation, we introduce a novel concept of channel-wise confidence in a node feature, which is assigned to each imputed channel feature of a node for reflecting certainty of the imputation. We then design pseudo-confidence using the channel-wise shortest path distance between a missing-feature node and its nearest known-feature node to replace unavailable true confidence in an actual learning process. Based on the pseudo-confidence, we propose a novel feature imputation scheme that performs channel-wise inter-node diffusion and node-wise inter-channel propagation. The scheme can endure even at an exceedingly high missing rate (e.g., 99.5\%) and it achieves state-of-the-art accuracy for both semi-supervised node classification and link prediction on various datasets containing a high rate of missing features. Codes are available at https://github.com/daehoum1/pcfi.

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。

↑