arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2310.14481cs.LGcs.SI
收藏

基于随机投影的高效异构图学习

Efficient Heterogeneous Graph Learning via Random Projection

School of Computing, National University of Singapore(新加坡国立大学计算机学院)

Jun Hu, Bryan Hooi, Bingsheng He

AI总结:

本文提出随机投影异构图神经网络RpHGNN,通过随机投影压缩和关系级邻居收集组件,在保持低信息损失的同时实现高效训练,在七个基准数据集上取得最优结果且速度提升230%。

摘要:

异构图神经网络(HGNNs)是在异构图上进行深度学习的强大工具。典型的HGNNs在训练过程中需要重复进行消息传递,这限制了其在大规模真实世界图上的效率。近期基于预计算的HGNNs使用一次性消息传递将异构图转换为规则形状的张量,从而实现高效的小批量训练。现有的基于预计算的HGNNs主要可分为两种风格,它们在允许的信息损失程度和效率上有所不同。我们提出了一种混合的基于预计算的HGNN,名为随机投影异构图神经网络(RpHGNN),它结合了一种风格的高效率和另一种风格的低信息损失优势。为了实现高效率,RpHGNN的主要框架由传播后更新的迭代组成,其中我们引入了一个随机投影压缩步骤,以确保复杂度仅线性增长。为了实现低信息损失,我们引入了一个关系级邻居收集组件,并采用奇偶传播方案,旨在以更细粒度的方式从邻居收集信息。实验结果表明,我们的方法在七个小型和大型基准数据集上取得了最先进的结果,同时与最有效的基线相比,速度提高了230%。令人惊讶的是,我们的方法不仅超越了基于预处理的基线,还优于端到端方法。

英文摘要:

Heterogeneous Graph Neural Networks (HGNNs) are powerful tools for deep learning on heterogeneous graphs. Typical HGNNs require repetitive message passing during training, limiting efficiency for large-scale real-world graphs. Recent pre-computation-based HGNNs use one-time message passing to transform a heterogeneous graph into regular-shaped tensors, enabling efficient mini-batch training. Existing pre-computation-based HGNNs can be mainly categorized into two styles, which differ in how much information loss is allowed and efficiency. We propose a hybrid pre-computation-based HGNN, named Random Projection Heterogeneous Graph Neural Network (RpHGNN), which combines the benefits of one style's efficiency with the low information loss of the other style. To achieve efficiency, the main framework of RpHGNN consists of propagate-then-update iterations, where we introduce a Random Projection Squashing step to ensure that complexity increases only linearly. To achieve low information loss, we introduce a Relation-wise Neighbor Collection component with an Even-odd Propagation Scheme, which aims to collect information from neighbors in a finer-grained way. Experimental results indicate that our approach achieves state-of-the-art results on seven small and large benchmark datasets while also being 230% faster compared to the most effective baseline. Surprisingly, our approach not only surpasses pre-processing-based baselines but also outperforms end-to-end methods.

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。

↑