arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2503.02448cs.LGcs.SI
收藏

NodeNAS:面向分布外泛化的节点特定图神经架构搜索

NodeNAS: Node-Specific Graph Neural Architecture Search for Out-of-Distribution Generalization

Tongji University(同济大学)

Qiyi Wang, Yinning Shao, Yunlong Ma, Min Liu

AI总结:

提出NodeNAS及多维扩展MNNAS方法,通过解耦节点拓扑与图分布、建模幂律分布,在有限数据下为不同节点定制架构,实现出色的分布外泛化。

摘要:

图神经架构搜索(GraphNAS)已展现出缓解图神经网络(GNN)因分布偏移导致性能下降的优势。近期方法在定制架构间引入权重共享,端到端地为每个图生成独特的GNN架构。然而,现有GraphNAS方法未考虑不同图间的分布模式,且严重依赖大量训练数据。在训练图稀疏或单一的情况下,这些方法难以发现图与架构间的最优映射,无法泛化到分布外(OOD)数据。本文提出节点特定图神经架构搜索(NodeNAS),旨在通过在有限数据集中解耦节点拓扑与图分布,为不同节点定制不同的聚合方法。我们进一步提出基于自适应聚合注意力的多维NodeNAS方法(MNNAS),学习具有良好泛化能力的节点特定架构定制器。具体而言,我们扩展了搜索空间的垂直深度,支持跨多个维度同时进行节点特定架构定制。此外,我们对不同同配性下节点度的幂律分布进行建模,编码结构不变信息以指导每个维度的架构定制。跨监督和无监督任务的大量实验表明,MNNAS超越了最先进算法并实现了出色的OOD泛化。

英文摘要:

Graph neural architecture search (GraphNAS) has demonstrated advantages in mitigating performance degradation of graph neural networks (GNNs) due to distribution shifts. Recent approaches introduce weight sharing across tailored architectures, generating unique GNN architectures for each graph end-to-end. However, existing GraphNAS methods do not account for distribution patterns across different graphs and heavily rely on extensive training data. With sparse or single training graphs, these methods struggle to discover optimal mappings between graphs and architectures, failing to generalize to out-of-distribution (OOD) data. In this paper, we propose node-specific graph neural architecture search(NodeNAS), which aims to tailor distinct aggregation methods for different nodes through disentangling node topology and graph distribution with limited datasets. We further propose adaptive aggregation attention based Multi-dim NodeNAS method(MNNAS), which learns an node-specific architecture customizer with good generalizability. Specifically, we extend the vertical depth of the search space, supporting simultaneous node-specific architecture customization across multiple dimensions. Moreover, we model the power-law distribution of node degrees under varying assortativity, encoding structure invariant information to guide architecture customization across each dimension. Extensive experiments across supervised and unsupervised tasks demonstrate that MNNAS surpasses state-of-the-art algorithms and achieves excellent OOD generalization.

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。

↑