用连续嵌入生成带类别图，避免了固定编码的几何缺陷。
Embedded Graph Flows for Categorical Graph Generation
- 用图变换器学习节点与边的连续嵌入，通过流模型从噪声中重构图结构。
- 在QM9上四项指标全最优，FCD达0.150，远优于基线方法。
- 适合分子结构生成，尤其擅长保持大分子局部子结构一致性。
生成带有类别的图需选择能构成连贯结构的节点和边类型，且不依赖节点顺序。现有方法将类别编码为固定的一热向量，可能导致类别间人为等距的几何假象。本文提出嵌入图流（Embedded Graph Flows, EGF），学习节点与无序边类别的连续嵌入，并利用置换等变图变换器将高斯噪声传输至这些学习到的终点。最终读出层将嵌入映射回离散图类别。在分子基准测试中，EGF表现优异：在QM9上，三项指标均优于对比方法，包括0.150的弗雷谢化学网络距离（FCD），显著低于分类扩散基线DiGress的0.717和基于桥接的基线GruM的0.812。在更大分子数据集ZINC250k上，采用邻域子图成对距离核（NSPDK）计算的最大均值差异（MMD）最低，表明其生成结果在局部子结构上与真实分子高度一致。代码已开源。
原文摘要 · Abstract (English)
Generating categorical graphs requires choosing node and edge types that form a coherent structure without depending on node order. Many graph generators encode categories as fixed one-hot vectors, which can impose an artificial geometry in which categories are equidistant. We propose Embedded Graph Flows (EGF), a generative model that learns continuous embeddings for node and unordered-edge categories and transports Gaussian noise towards these learnt endpoints using a permutation-equivariant graph transformer. A terminal readout maps the embeddings back to discrete graph categories. Across molecular benchmarks, EGF achieved competitive performance. On QM9, EGF gives the best result on all four reported metrics among the three methods, including a Fr\'echet ChemNet Distance (FCD) of 0.150, compared with 0.717 for the categorical-diffusion baseline DiGress and 0.812 for the bridge-based baseline GruM. When applied to larger molecules in ZINC250k, EGF retains the lowest maximum mean discrepancy (MMD) using the neighbourhood subgraph pairwise distance kernel (NSPDK), indicating close agreement with the local substructures of the reference molecules. Our code is available at https://github.com/Trusted-System-Lab/EGF.
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。