A noise injection strategy for graph autoencoder training
A noise injection strategy for graph autoencoder training
复制标题
DOI：
10.1007/s00521-020-05283-x
复制
发表时间：
2020-08
影响因子：
6
通讯作者：
Yingfeng Wang;Biyun Xu;M. Kwak;Xiaoqin Zeng
中科院分区：
文献类型：
--
作者：
Yingfeng Wang;Biyun Xu;M. Kwak;Xiaoqin Zeng
图形自动编码器可以将图形数据映射到低维空间中。它是一种用于图分析的强大的图嵌入方法，可以降低计算量。研究人员开发了不同的图形自编码器来满足不同的需求。提出了一种基于噪声注入的图自编码器训练策略。这是一种通用的训练策略，可以灵活地适应大多数现有的训练算法。实验结果验证了这种通用策略可以显著减少过拟合，并识别噪声率设置，以一致地提高训练性能。
Graph autoencoder can map graph data into a low-dimensional space. It is a powerful graph embedding method applied in graph analytics to lower the computational cost. Researchers have developed different graph autoencoders for addressing different needs. This paper proposes a strategy based on noise injection for graph autoencoder training. This is a general training strategy that can flexibly fit most existing training algorithms. The experimental results verify this general strategy can significantly reduce overfitting and identify the noise rate setting for consistent training performance improvement.