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 2302.08990cs.LGcs.AI
收藏

通过投影有效地忘记你在图表示学习中学到的东西

Efficiently Forgetting What You Have Learned in Graph Representation Learning via Projection

The Pennsylvania State University(宾夕法尼亚州立大学)

Weilin Cong, Mehrdad Mahdavi

摘要:

随着隐私保护受到越来越多的关注,从预训练的图学习模型中消除特定节点的影响也变得同样重要。然而,由于图结构数据中的节点依赖性,图神经网络(GNNs)中的表示学习是具有挑战性的,并且还没有得到很好的探索。在本文中,我们填补了这一空白,首先研究了线性GNNs中的非学习问题,然后将其扩展到非线性结构。给定一组要遗忘的节点,我们提出了一种投影器,它通过将预训练模型的权重参数投影到与要遗忘的节点的特征无关的子空间上来进行遗忘。PROJECTOR可以克服节点依赖性带来的挑战,并享受完美的数据删除,即,未学习的模型参数不包含关于由算法构造保证的未学习的节点特征的任何信息。在真实数据集上的实验结果表明了PROJECTOR的有效性和效率。

英文摘要:

As privacy protection receives much attention, unlearning the effect of a specific node from a pre-trained graph learning model has become equally important. However, due to the node dependency in the graph-structured data, representation unlearning in Graph Neural Networks (GNNs) is challenging and less well explored. In this paper, we fill in this gap by first studying the unlearning problem in linear-GNNs, and then introducing its extension to non-linear structures. Given a set of nodes to unlearn, we propose PROJECTOR that unlearns by projecting the weight parameters of the pre-trained model onto a subspace that is irrelevant to features of the nodes to be forgotten. PROJECTOR could overcome the challenges caused by node dependency and enjoys a perfect data removal, i.e., the unlearned model parameters do not contain any information about the unlearned node features which is guaranteed by algorithmic construction. Empirical results on real-world datasets illustrate the effectiveness and efficiency of PROJECTOR.

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。

↑