SUGAR: Efficient Subgraph-Level Training via Resource-Aware Graph Partitioning
SUGAR: Efficient Subgraph-Level Training via Resource-Aware Graph Partitioning
复制标题
DOI：
10.1109/tc.2023.3288755
复制
发表时间：
2022-01
影响因子：
3.7
通讯作者：
Zihui Xue;Yuedong Yang-;Mengtian Yang;R. Marculescu
中科院分区：
文献类型：
--
作者：
Zihui Xue;Yuedong Yang-;Mengtian Yang;R. Marculescu
图神经网络（gnn）在各种基于图的应用中显示出巨大的潜力，例如推荐系统、药物发现和对象识别。然而，资源高效的GNN学习是一个很少被探索的话题，尽管它对边缘计算和物联网（IoT）应用有很多好处。为了改善这种状况，本工作提出了通过资源感知图划分（SUGAR）进行有效的子图级训练。SUGAR首先将初始图划分为一组不相交的子图，然后在子图层面进行局部训练。我们提供了理论分析，并在五个图基准上进行了广泛的实验，以验证其在实践中的有效性。我们在五种不同硬件平台上的结果表明，在大规模图形上使用SUGAR可以极大地提高运行时速度并减少内存。我们相信SUGAR为开发资源高效的GNN方法开辟了一个新的研究方向，因此适合物联网部署。
Graph Neural Networks (GNNs) have demonstrated a great potential in a variety of graph-based applications, such as recommender systems, drug discovery, and object recognition. Nevertheless, resource-efficient GNN learning is a rarely explored topic despite its many benefits for edge computing and Internet of Things (IoT) applications. To improve this state of affairs, this work proposes efficient subgraph-level training via resource-aware graph partitioning (SUGAR). SUGAR first partitions the initial graph into a set of disjoint subgraphs and then performs local training at the subgraph-level We provide a theoretical analysis and conduct extensive experiments on five graph benchmarks to verify its efficacy in practice. Our results across five different hardware platforms demonstrate great runtime speedup and memory reduction of SUGAR on large-scale graphs. We believe SUGAR opens a new research direction towards developing GNN methods that are resource-efficient, hence suitable for IoT deployment.