Discerning Edge Influence for Network Embedding
Discerning Edge Influence for Network Embedding
复制标题
DOI：
10.1145/3357384.3358044
复制
发表时间：
2019-11
期刊：
影响因子：
--
通讯作者：
Yaojing Wang;Yuan Yao;Hanghang Tong;F. Xu;Jian Lu
中科院分区：
文献类型：
--
作者：
Yaojing Wang;Yuan Yao;Hanghang Tong;F. Xu;Jian Lu
网络嵌入，学习节点的低维表示，已经获得了显着的研究关注。尽管它的上级经验成功，通常由下游任务的预测性能来衡量（例如，多标签分类），不清楚为什么给定的嵌入算法输出特定的节点表示，以及所得到的节点表示如何与输入网络的结构相关。在本文中，我们建议识别边缘的影响，作为理解skip-gram basd网络嵌入方法的第一步。为此，我们提出了一个审计框架近，其关键部分包括两个算法（近添加\和近删除），以有效和高效地量化每个边缘的影响。基于该算法，我们进一步确定高影响力的边缘，通过利用边缘的影响力和网络结构之间的联系。实验结果表明，所提出的算法（Near-add \和Near-del）是显着更快（高达2,000\times $）比简单的方法，几乎没有质量损失。此外，所提出的框架可以有效地识别最有影响力的网络嵌入在下游预测任务和对抗性攻击的背景下的边缘。
Network embedding, which learns the low-dimensional representations of nodes, has gained significant research attention. Despite its superior empirical success, often measured by the prediction performance of downstream tasks (e.g., multi-label classification), it is unclear \em why a given embedding algorithm outputs the specific node representations, and \em how the resulting node representations relate to the structure of the input network. In this paper, we propose to discern the edge influence as the first step towards understanding skip-gram basd network embedding methods. For this purpose, we propose an auditing framework Near, whose key part includes two algorithms (Near-add \ and Near-del ) to effectively and efficiently quantify the influence of each edge. Based on the algorithms, we further identify high-influential edges by exploiting the linkage between edge influence and the network structure. Experimental results demonstrate that the proposed algorithms (Near-add \ and Near-del ) are significantly faster (up to $2,000\times$) than straightforward methods with little quality loss. Moreover, the proposed framework can efficiently identify the most influential edges for network embedding in the context of downstream prediction task and adversarial attacking.