发表机构
TU Delft; University of Seville; University of Barcelona; Harvard University; ICREA（代尔夫特理工大学; 塞维利亚大学; 巴塞罗那大学; 哈佛大学; 加泰罗尼亚研究与高级研究院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究提出零样本迁移协议，通过几何重归一化缩小图训练GNN，可直接将权重迁移到全分辨率图，保留预测性能并降低训练成本，为尺度等变图架构提供了方向。
AI 中文摘要
图神经网络（GNN）可在大图上运行，但在数百万节点规模时对基础设施敏感，即便处理更大图也通常需要可扩展的训练技术，这引出核心问题：在图的缩小副本上训练的模型何时能直接部署到全分辨率图上而无需重新训练？我们提出一种零样本迁移协议，其中GNN在经几何重归一化（GR）粗粒化的图上训练，所得权重直接迁移到原始网络。在合成及真实网络中，在GR缩小副本上训练保留了大部分原始规模的预测性能，同时显著降低训练成本。我们进一步发现，学习到的表示和预测轨迹在不同尺度间仍保持对齐。这些结果表明，结构相似性可能比网络规模更能决定GNN的可迁移性，为尺度等变图架构开辟了路径。
英文摘要
Graph neural networks (GNNs) can operate on large graphs but become infrastructure-sensitive at the scale of millions of nodes and typically require scalable training techniques for even larger graphs. This raises a central question: when can a model trained on a smaller, scaled-down replica of a graph be deployed on the full-resolution graph without retraining? We introduce a zero-shot transfer protocol in which a GNN is trained on a graph coarse-grained by geometric renormalization (GR), and the resulting weights are transferred directly to the original network. Across synthetic and real-world networks, training on GR scaled-down replicas preserves much of the original-scale predictive performance while significantly reducing training cost. We further find that learned representations and predictive trajectories remain aligned across scales. These findings suggest that structural similarity may be more important than network size in determining GNN transferability, opening a path toward scale-equivariant graph architectures.
