NodeNAS: Node-Specific Graph Neural Architecture Search for Out-of-Distribution Generalization
This addresses a domain-specific problem for graph learning researchers and practitioners, offering an incremental improvement over existing GraphNAS methods by enhancing OOD generalization with sparse data.
The paper tackles the problem of graph neural architecture search (GraphNAS) failing to generalize to out-of-distribution data with limited training graphs, by proposing NodeNAS and MNNAS methods that tailor architectures per node using topology and distribution disentanglement, achieving state-of-the-art performance in supervised and unsupervised tasks.
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.