用对称性设计噪声，让图神经网络更通用更强。
Using Random Noise Equivariantly to Boost Graph Neural Networks Universally
- 利用噪声的对称性设计新架构，避免性能下降。
- 在节点、链接、子图和图级别任务中显著提升效果。
- 无需针对特定任务调整，适合各类图学习场景。
近期图神经网络（GNN）研究探索了将随机噪声作为输入特征以增强模型表达能力的潜力。然而，盲目引入噪声会降低性能，而针对特定任务设计的架构虽表现优异却缺乏普适性。本文建立理论框架，揭示未精心设计时引入噪声会增加样本复杂度。为此提出等变噪声GNN（ENGNN），通过利用噪声的对称性质缓解样本复杂度问题，提升泛化能力。实验表明，等变地使用噪声可显著改善节点级、链接级、子图级和图级任务的表现，且性能媲美专用模型，为多种图任务提供通用的表达力增强方法。
原文摘要 · Abstract (English)
Recent advances in Graph Neural Networks (GNNs) have explored the potential of random noise as an input feature to enhance expressivity across diverse tasks. However, naively incorporating noise can degrade performance, while architectures tailored to exploit noise for specific tasks excel yet lack broad applicability. This paper tackles these issues by laying down a theoretical framework that elucidates the increased sample complexity when introducing random noise into GNNs without careful design. We further propose Equivariant Noise GNN (ENGNN), a novel architecture that harnesses the symmetrical properties of noise to mitigate sample complexity and bolster generalization. Our experiments demonstrate that using noise equivariantly significantly enhances performance on node-level, link-level, subgraph, and graph-level tasks and achieves comparable performance to models designed for specific tasks, thereby offering a general method to boost expressivity across various graph tasks.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。