当前位置: X-MOL 学术arXiv.cs.LG › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Adversarial Learning for Debiasing Knowledge Graph Embeddings
arXiv - CS - Machine Learning Pub Date : 2020-06-29 , DOI: arxiv-2006.16309
Mario Arduini, Lorenzo Noci, Federico Pirovano, Ce Zhang, Yash Raj Shrestha, Bibek Paudel

Knowledge Graphs (KG) are gaining increasing attention in both academia and industry. Despite their diverse benefits, recent research have identified social and cultural biases embedded in the representations learned from KGs. Such biases can have detrimental consequences on different population and minority groups as applications of KG begin to intersect and interact with social spheres. This paper aims at identifying and mitigating such biases in Knowledge Graph (KG) embeddings. As a first step, we explore popularity bias -- the relationship between node popularity and link prediction accuracy. In case of node2vec graph embeddings, we find that prediction accuracy of the embedding is negatively correlated with the degree of the node. However, in case of knowledge-graph embeddings (KGE), we observe an opposite trend. As a second step, we explore gender bias in KGE, and a careful examination of popular KGE algorithms suggest that sensitive attribute like the gender of a person can be predicted from the embedding. This implies that such biases in popular KGs is captured by the structural properties of the embedding. As a preliminary solution to debiasing KGs, we introduce a novel framework to filter out the sensitive attribute information from the KG embeddings, which we call FAN (Filtering Adversarial Network). We also suggest the applicability of FAN for debiasing other network embeddings which could be explored in future work.

中文翻译:

用于消除知识图谱嵌入的对抗性学习

知识图谱 (KG) 在学术界和工业界都越来越受到关注。尽管它们的好处多种多样,但最近的研究已经确定了从 KG 学习的表征中嵌入的社会和文化偏见。随着 KG 的应用开始与社会领域交叉和互动,这种偏见会对不同的人口和少数群体产生不利影响。本文旨在识别和减轻知识图谱 (KG) 嵌入中的此类偏差。作为第一步,我们探索流行度偏差——节点流行度和链接预测准确性之间的关系。在 node2vec 图嵌入的情况下,我们发现嵌入的预测精度与节点的度数呈负相关。然而,在知识图嵌入(KGE)的情况下,我们观察到相反的趋势。作为第二步,我们探索了 KGE 中的性别偏见,对流行的 KGE 算法的仔细检查表明,可以从嵌入中预测诸如人的性别之类的敏感属性。这意味着流行的 KG 中的这种偏差是由嵌入的结构特性捕获的。作为消除 KG 的初步解决方案,我们引入了一种新颖的框架来过滤掉 KG 嵌入中的敏感属性信息,我们称之为 FAN(过滤对抗网络)。我们还建议 FAN 适用于消除其他网络嵌入的偏差,这可以在未来的工作中进行探索。这意味着流行的 KG 中的这种偏差是由嵌入的结构特性捕获的。作为消除 KG 的初步解决方案,我们引入了一种新颖的框架来过滤掉 KG 嵌入中的敏感属性信息,我们称之为 FAN(过滤对抗网络)。我们还建议 FAN 适用于消除其他网络嵌入的偏差,这可以在未来的工作中进行探索。这意味着流行的 KG 中的这种偏差是由嵌入的结构特性捕获的。作为消除 KG 的初步解决方案,我们引入了一种新颖的框架来过滤掉 KG 嵌入中的敏感属性信息,我们称之为 FAN(过滤对抗网络)。我们还建议 FAN 适用于消除其他网络嵌入的偏差,这可以在未来的工作中进行探索。
更新日期:2020-07-01
down
wechat
bug