提出噪声条件图网络，让生成图时的邻居信息范围和分辨率随噪声动态调整。
Geometric Generative Modeling with Noise-Conditioned Graph Networks
- 根据噪声水平动态调整消息传递范围与图分辨率
- 在3D点云、时空转录组等数据上显著优于传统模型
- 适合需要生成带空间结构图的科研与工程应用
具有空间结构的图生成在计算机图形学和空间基因组学等领域至关重要。近期基于流的生成模型通过逐步加噪再学习去噪实现优异效果。然而现有模型使用与噪声水平无关的图神经网络架构，限制了表达能力。为此，我们提出噪声条件图网络（NCGNs），其架构随生成过程中的噪声水平动态变化。理论与实证分析表明：随着噪声增加，(1) 图需要来自更远邻域的信息，(2) 可在更低分辨率下有效表示。基于此，我们设计动态消息传递（DMP）作为NCGNs的具体实例，自适应调整消息传递范围与分辨率。DMP在3D点云、时空转录组及图像等多个领域均持续优于独立于噪声的架构。代码已开源。
原文摘要 · Abstract (English)
Generative modeling of graphs with spatial structure is essential across many applications from computer graphics to spatial genomics. Recent flow-based generative models have achieved impressive results by gradually adding and then learning to remove noise from these graphs. Existing models, however, use graph neural network architectures that are independent of the noise level, limiting their expressiveness. To address this issue, we introduce \textit{Noise-Conditioned Graph Networks} (NCGNs), a class of graph neural networks that dynamically modify their architecture according to the noise level during generation. Our theoretical and empirical analysis reveals that as noise increases, (1) graphs require information from increasingly distant neighbors and (2) graphs can be effectively represented at lower resolutions. Based on these insights, we develop Dynamic Message Passing (DMP), a specific instantiation of NCGNs that adapts both the range and resolution of message passing to the noise level. DMP consistently outperforms noise-independent architectures on a variety of domains including $3$D point clouds, spatiotemporal transcriptomics, and images. Code is available at https://github.com/peterpaohuang/ncgn.
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。