arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2412.02196cs.LG
收藏

SA-GNAS:面向高效大规模图神经架构搜索的种子架构扩展

SA-GNAS: Seed Architecture Expansion for Efficient Large-scale Graph Neural Architecture Search

State Key Laboratory for Novel Software Technology, Nanjing University(南京大学 现代软件工程国家重点实验室)

Guanghui Zhu, Zipeng Ji, Jingyan Chen, Limin Wang, Chunfeng Yuan, Yihua Huang

AI总结:

针对大规模图神经架构搜索成本高的问题,提出SA-GNAS,通过性能排序一致性选择种子架构并基于熵最小化迭代扩展,在五个大规模图上提升性能与搜索效率。

摘要:

GNAS(图神经架构搜索)已在为节点分类和链接预测等多个下游任务自动设计最优图神经架构方面展现出显著效果。然而,由于高昂的计算和内存开销,大多数现有GNAS方法无法高效处理包含百万级以上节点和边的大规模图。为了在大规模图上扩展GNAS并取得更优性能,我们提出SA-GNAS,这是一种基于种子架构扩展的新型高效大规模GNAS框架。类似于生物技术中的细胞扩增,我们首先构建一个种子架构,然后迭代扩展该种子架构。具体而言,我们首先提出一种基于性能排序一致性的种子架构选择方法,用于选择在子图上搜索到的、与原始大规模图最匹配的架构。随后,我们提出一种基于熵最小化的种子架构扩展方法,以进一步提升种子架构的性能。在五个大规模图上的大量实验结果表明,所提出的SA-GNAS优于人工设计的最先进GNN架构以及现有的图NAS方法。此外,SA-GNAS能够显著减少搜索时间,表现出更好的搜索效率。对于拥有十亿条边的最大图,SA-GNAS相较于SOTA大规模GNAS方法GAUSS可实现2.8倍加速。另外,由于SA-GNAS本身具有并行化特性,使用更多GPU可以进一步提高搜索效率。SA-GNAS可在https://github.com/PasaLab/SAGNAS获取。

英文摘要:

GNAS (Graph Neural Architecture Search) has demonstrated great effectiveness in automatically designing the optimal graph neural architectures for multiple downstream tasks, such as node classification and link prediction. However, most existing GNAS methods cannot efficiently handle large-scale graphs containing more than million-scale nodes and edges due to the expensive computational and memory overhead. To scale GNAS on large graphs while achieving better performance, we propose SA-GNAS, a novel framework based on seed architecture expansion for efficient large-scale GNAS. Similar to the cell expansion in biotechnology, we first construct a seed architecture and then expand the seed architecture iteratively. Specifically, we first propose a performance ranking consistency-based seed architecture selection method, which selects the architecture searched on the subgraph that best matches the original large-scale graph. Then, we propose an entropy minimization-based seed architecture expansion method to further improve the performance of the seed architecture. Extensive experimental results on five large-scale graphs demonstrate that the proposed SA-GNAS outperforms human-designed state-of-the-art GNN architectures and existing graph NAS methods. Moreover, SA-GNAS can significantly reduce the search time, showing better search efficiency. For the largest graph with billion edges, SA-GNAS can achieve 2.8 times speedup compared to the SOTA large-scale GNAS method GAUSS. Additionally, since SA-GNAS is inherently parallelized, the search efficiency can be further improved with more GPUs. SA-GNAS is available at https://github.com/PasaLab/SAGNAS.

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。

↑