arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2008.07672cs.LGstat.ML
收藏

基于张量分解的包围节点嵌入--以DeepWalk为例

Ensemble Node Embeddings using Tensor Decomposition: A Case-Study on DeepWalk

University of Texas Rio Grande Valley(德克萨斯大学里奥格兰德河谷分校);University of California Riverside(加利福尼亚大学河滨分校)

Jia Chen, Evangelos E. Papalexakis

摘要:

在过去的几年里,节点嵌入引起了越来越多的关注。在这种情况下,我们提出了一种新的集成节点嵌入方法,称为TenSemble 2 Vec,首先使用现有技术生成多个嵌入,并将它们作为最先进的张量分解模型(即PARAFAC 2)的多视图数据输入,以学习节点的共享低维表示。与其他嵌入方法相反,我们的TenSemble 2 Vec利用了不同方法或具有不同超参数的相同方法的互补信息,从而绕过了选择模型的挑战。大量的测试使用真实世界的数据验证了所提出的方法的效率。

英文摘要:

Node embeddings have been attracting increasing attention during the past years. In this context, we propose a new ensemble node embedding approach, called TenSemble2Vec, by first generating multiple embeddings using the existing techniques and taking them as multiview data input of the state-of-art tensor decomposition model namely PARAFAC2 to learn the shared lower-dimensional representations of the nodes. Contrary to other embedding methods, our TenSemble2Vec takes advantage of the complementary information from different methods or the same method with different hyper-parameters, which bypasses the challenge of choosing models. Extensive tests using real-world data validates the efficiency of the proposed method.

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。

↑