arXiv：cs.LG（机器学习，全量分类）· Umer Gupta, Saku Peltonen, Martin Ritzert·· 5 小时前AI 评分26
Autoregressive Frontier Expansion：用图机器学习生成树状分支结构
Autoregressive Frontier Expansion: Growing Trees with Graph Machine Learning
AI 导读
研究者提出 Autoregressive Frontier Expansion，一种通过迭代扩展过程构建树状结构的生成框架，模拟真实树木的生物学生长。每一步由 SO(2)-equivariant GNN 参数化的 flow-matching 模型预测各活跃分支是分叉还是终止，从而扩展前沿。该方法在皮层神经元和植物树的非条件、类别条件及形态引导生成任务中，生成的形态与参考分布高度吻合。
正文
Abstract:Tree-like branching structures are common in nature, from botanical trees to neurons, blood vessels and respiratory trees. Their branching shape often reflects function, making structural modelling central to understanding how these systems work. Because acquiring real-world 3D data is often expensive or infeasible, realistic generative models are valuable for simulation and data augmentation. Existing morphology-specific models either constrain how topology is generated or rely on hand-tuned, mechanistic procedures. Generic 3D graph generators, by contrast, do not exploit or enforce the structure of trees. We propose Autoregressive Frontier Expansion, a generative framework that constructs trees through an iterative expansion process, simulating the biological growth of real trees. At each step, a flow-matching model parameterised by an SO(2)-equivariant GNN expands the frontier by predicting whether each active branch bifurcates or terminates. We evaluate our method on cortical neurons and botanical trees in unconditional, class-conditioned, and morphology-guided generation. Across both domains, the generated morphologies agree closely with the reference distributions and, in conditional experiments, with the specified targets.
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2609.38506 [cs.LG] |
| (or arXiv:2609.38506v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2609.38506 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Saku Peltonen [view email]
[v1]
Tue, 29 Sep 2026 20:29:07 UTC (9,815 KB)
来源：arXiv：cs.LG（机器学习，全量分类） · arxiv.org