arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2111.06345cs.LGcs.AIcs.CLcs.NE
收藏

基于关系推理模式的中毒知识图谱嵌入

Poisoning Knowledge Graph Embeddings via Relation Inference Patterns

Trinity College Dublin(都柏林三一学院);TU Dublin(都柏林理工大学);Accenture Labs(埃森哲实验室)

Peru Bhardwaj, John Kelleher, Luca Costabello, Declan O'Sullivan

摘要:

针对知识图谱中的链接预测任务,研究了针对知识图谱嵌入模型的数据投毒攻击问题。为了毒化KGE模型,我们建议利用它们的归纳能力,这些能力是通过知识图谱中的对称、反转和组合等关系模式捕获的。具体来说,降低模型的预测信心的目标事实,我们建议提高模型的预测信心的一组诱饵事实。因此,我们制作了对抗性添加,可以通过不同的推理模式提高模型对诱饵事实的预测信心。我们的实验表明,提出的中毒攻击优于最先进的基线上的四个KGE模型的两个公开可用的数据集。我们还发现,基于对称模式的攻击概括了所有的模型-数据集组合,这表明KGE模型对这种模式的敏感性。

英文摘要:

We study the problem of generating data poisoning attacks against Knowledge Graph Embedding (KGE) models for the task of link prediction in knowledge graphs. To poison KGE models, we propose to exploit their inductive abilities which are captured through the relationship patterns like symmetry, inversion and composition in the knowledge graph. Specifically, to degrade the model's prediction confidence on target facts, we propose to improve the model's prediction confidence on a set of decoy facts. Thus, we craft adversarial additions that can improve the model's prediction confidence on decoy facts through different inference patterns. Our experiments demonstrate that the proposed poisoning attacks outperform state-of-art baselines on four KGE models for two publicly available datasets. We also find that the symmetry pattern based attacks generalize across all model-dataset combinations which indicates the sensitivity of KGE models to this pattern.

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。

↑