arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2201.09391cs.LGcs.AI
收藏

基于划分的图神经网络主动学习

Partition-Based Active Learning for Graph Neural Networks

University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校);University of Michigan(密歇根大学)

Jiaqi Ma, Ziqiao Ma, Joyce Chai, Qiaozhu Mei

摘要:

我们研究了半监督学习问题与图神经网络(GNNs)在主动学习设置。我们提出了GraphPart,一种新的基于分区的主动学习方法的GNNs。GraphPart首先将图分割成不相交的分区,然后在每个分区中选择代表节点进行查询。该方法的动机是一个新的分析的分类错误下现实的平滑假设的图形和节点功能。在多个基准数据集上进行的大量实验表明,该方法在广泛的注释预算约束下优于现有的GNN主动学习方法。此外,所提出的方法不引入额外的超参数,这对于模型训练至关重要,特别是在标记验证集可能不可用的主动学习设置中。

英文摘要:

We study the problem of semi-supervised learning with Graph Neural Networks (GNNs) in an active learning setup. We propose GraphPart, a novel partition-based active learning approach for GNNs. GraphPart first splits the graph into disjoint partitions and then selects representative nodes within each partition to query. The proposed method is motivated by a novel analysis of the classification error under realistic smoothness assumptions over the graph and the node features. Extensive experiments on multiple benchmark datasets demonstrate that the proposed method outperforms existing active learning methods for GNNs under a wide range of annotation budget constraints. In addition, the proposed method does not introduce additional hyperparameters, which is crucial for model training, especially in the active learning setting where a labeled validation set may not be available.

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。

↑