Towards Fair Representation Learning in Knowledge Graph with Stable Adversarial Debiasing
Towards Fair Representation Learning in Knowledge Graph with Stable Adversarial Debiasing
复制标题
DOI：
10.1109/icdmw58026.2022.00119
复制
发表时间：
2022-11
期刊：
影响因子：
--
通讯作者：
Yihe Wang;Mohammad Mahdi Khalili;X. Zhang
中科院分区：
文献类型：
--
作者：
Yihe Wang;Mohammad Mahdi Khalili;X. Zhang
通过图形结构的巨大信息，知识图（kg）引起了对ACA流行研究和工业应用的日益兴趣。最近的研究表明，就敏感属性（例如性别和种族）而言，人口偏见存在于KG实体的学习代表中。这种偏见会对特定的人口产生负面影响，尤其是少数民族和代表性不足的群体，并加剧基于机器学习的人类不平等。通过同时训练特定于任务的预测指标和敏感属性特定的歧视器，对抗性学习被认为是减轻表示学习模型中偏见的有效方法。但是，由于拓扑结构造成的独特挑战和知识实体之间的全面重新统治，因此在知识图中的表示学习中很少研究基于对抗性的学习偏差。在本文中，我们提出了一个框架，以学习知识图挖掘中节点和边缘的无偏表示。具体而言，我们将一种简单但有效的标准化技术与图形神经网络（GNN）集成在一起，以限制权重更新过程。此外，作为一份正在进行的纸张，我们还发现，引入的权重标准化技术可以减轻对抗性贬低对公平和稳定的机器学习的不稳定性的陷阱。我们在具有多种边缘类型和节点类型的基准图上评估了提出的框架。实验结果表明，我们的模型在三个竞争基准的基准方面实现了可比性或更好的性别公平性。重要的是，我们在公平模型中的优越性不会吓到知识图任务（即多级边缘分类）中的性能。
With graph-structured tremendous information, Knowledge Graphs (KG) aroused increasing interest in aca-demic research and industrial applications. Recent studies have shown demographic bias, in terms of sensitive attributes (e.g., gender and race), exist in the learned representations of KG entities. Such bias negatively affects specific popu-lations, especially minorities and underrepresented groups, and exacerbates machine learning-based human inequality. Adversariallearning is regarded as an effective way to alleviate bias in the representation learning model by simultaneously training a task-specific predictor and a sensitive attribute-specific discriminator. However, due to the unique challenge caused by topological structure and the comprehensive re-lationship between knowledge entities, adversarial learning-based debiasing is rarely studied in representation learning in knowledge graphs. In this paper, we propose a framework to learn unbiased representations for nodes and edges in knowledge graph mining. Specifically, we integrate a simple-but-effective normalization technique with Graph Neural Networks (GNNs) to constrain the weights updating process. Moreover, as a work-in-progress paper, we also find that the introduced weights normalization technique can mitigate the pitfalls of instability in adversarial debasing towards fair-and-stable machine learning. We evaluate the proposed framework on a benchmarking graph with multiple edge types and node types. The experimental results show that our model achieves comparable or better gender fairness over three competitive baselines on Equality of Odds. Importantly, our superiority in the fair model does not scarify the performance in the knowledge graph task (i.e., multi-class edge classification).