arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2005.04081cs.LGcs.SIphysics.soc-phstat.ML
收藏

利用图卷积网络从数据中提取几何图以辅助分类任务

Geometric graphs from data to aid classification tasks with graph convolutional networks

Queen Mary University of London(伦敦玛丽女王大学);Imperial College London(帝国理工学院);Tokyo Institute of Technology(东京工业大学)

Yifan Qian, Paul Expert, Pietro Panzarasa, Mauricio Barahona

摘要:

传统的分类任务学习仅基于样本特征将样本分配给给定的类。这一范式正在演变,以包括其他信息来源,如样本之间的已知关系。在这里,我们表明,即使额外的关系信息是不可用的数据集,可以通过构建几何图形的功能本身,并使用它们在一个图卷积网络的分类。通过以相对低的边缘密度捕获样本相似性的图来最大化分类精度的提高。我们表明,这种特征衍生图增加了数据与地面实况的对齐,同时提高了类分离。我们还证明了使用谱稀疏化可以使图更有效,这减少了边缘的数量,同时仍然提高了分类性能。我们使用来自各个科学领域的合成和真实世界的数据集来说明我们的研究结果。

英文摘要:

Traditional classification tasks learn to assign samples to given classes based solely on sample features. This paradigm is evolving to include other sources of information, such as known relations between samples. Here we show that, even if additional relational information is not available in the data set, one can improve classification by constructing geometric graphs from the features themselves, and using them within a Graph Convolutional Network. The improvement in classification accuracy is maximized by graphs that capture sample similarity with relatively low edge density. We show that such feature-derived graphs increase the alignment of the data to the ground truth while improving class separation. We also demonstrate that the graphs can be made more efficient using spectral sparsification, which reduces the number of edges while still improving classification performance. We illustrate our findings using synthetic and real-world data sets from various scientific domains.

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。

↑