arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1809.05343cs.CV
收藏

面向快速图表示学习的自适应采样

Adaptive Sampling Towards Fast Graph Representation Learning

Tencent AI Lab(腾讯人工智能实验室);Australian National University(澳大利亚国立大学)

Wenbing Huang, Tong Zhang, Yu Rong, Junzhou Huang

摘要:

图卷积网络已经成为学习图的顶点表示的重要工具。在大规模图上适配GCN的主要挑战是可扩展性问题,由于不可控的邻域扩展,它在计算和存储方面都产生了巨大的成本。在本文中,我们通过开发一种自适应分层采样方法来加速GCN的训练。通过自顶向下逐层构造网络,以上层节点为条件对下层节点进行采样,采样邻域由不同的父结点共享,由于采样大小固定,避免了网络的过度膨胀。更重要的是,所提出的采样器是自适应的,适用于显式方差减少,这反过来又增强了我们的方法的训练。此外,我们提出了一种新颖且经济的方法,通过应用跳过连接来促进消息在远程节点上传递。在多个基准上的实验验证了该方法在提高分类精度的同时具有更快的收敛速度。

英文摘要:

Graph Convolutional Networks (GCNs) have become a crucial tool on learning representations of graph vertices. The main challenge of adapting GCNs on large-scale graphs is the scalability issue that it incurs heavy cost both in computation and memory due to the uncontrollable neighborhood expansion across layers. In this paper, we accelerate the training of GCNs through developing an adaptive layer-wise sampling method. By constructing the network layer by layer in a top-down passway, we sample the lower layer conditioned on the top one, where the sampled neighborhoods are shared by different parent nodes and the over expansion is avoided owing to the fixed-size sampling. More importantly, the proposed sampler is adaptive and applicable for explicit variance reduction, which in turn enhances the training of our method. Furthermore, we propose a novel and economical approach to promote the message passing over distant nodes by applying skip connections. Intensive experiments on several benchmarks verify the effectiveness of our method regarding the classification accuracy while enjoying faster convergence speed.

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。

↑