Learning Fair Graph Representations via Automated Data Augmentations-猫眼课题宝

DOI：

发表时间：

2023

期刊：

影响因子：

通讯作者：

中科院分区：

其他

文献类型：

作者：

Hongyi Ling;Zhimeng Jiang;Youzhi Luo;S. Ji;Na Zou

我们考虑通过数据增强来学习公平的图表学习。虽然先前已经探索过这个方向，但现有方法总是依赖于公平图数据属性的某些假设，以便在数据增强方面设计固定的策略。然而，在不同的情况下，公平图数据的确切属性可能会有很大差异。因此，在不同的应用程序方案中，启发式设计的增强可能并不总是会生成公平的图形数据。在这项工作中，我们提出了一种称为Graphair的方法，以基于自动图数据增强来学习公平表示。这种公平意识的增强本身就是从数据中学到的。我们的Graphair旨在自动从输入图中发现公平感知到的增强，以规避敏感信息，同时保留其他有用的信息。实验结果表明，就公平性 - 准确性权衡绩效而言，我们的Graphair始终在多个节点分类数据集上胜过许多基准。此外，结果表明，Graphair可以自动学习生成公平的图形数据，而无需事先了解与公平相关的图形属性。我们的代码是DIG软件包（https://github.com/divelab/dig）的一部分。

We consider fair graph representation learning via data augmentations. While this direction has been explored previously, existing methods invariably rely on certain assumptions on the properties of fair graph data in order to design fixed strategies on data augmentations. Nevertheless, the exact properties of fair graph data may vary significantly in different scenarios. Hence, heuristically designed augmentations may not always generate fair graph data in different application scenarios. In this work, we propose a method, known as Graphair, to learn fair representations based on automated graph data augmentations. Such fairness-aware augmentations are themselves learned from data. Our Graphair is designed to automatically discover fairness-aware augmentations from input graphs in order to circumvent sensitive information while preserving other useful information. Experimental results demonstrate that our Graphair consistently outperforms many baselines on multiple node classification datasets in terms of fairness-accuracy trade-off performance. In addition, results indicate that Graphair can automatically learn to generate fair graph data without prior knowledge on fairness-relevant graph properties. Our code is publicly available as part of the DIG package (https://github.com/divelab/DIG).

Learning Fair Graph Representations via Automated Data Augmentations

文献摘要

相似文献