发表机构
McWilliams School of Biomedical Informatics; University of Texas Health Science Center at Houston（麦克威廉姆斯生物医学信息学院; 德克萨斯大学休斯顿健康科学中心）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
EvoTreeNAD提出谱系引导进化算法，无需种子或搜索空间，通过前百分位选择与智能体协作，发现超越基线的神经架构，在CIFAR和MedMNIST上表现优异。
AI 中文摘要
AI驱动的科学发现通过自主开发解决方案和设计来加速研究。大型语言模型（LLM）智能体通过迭代生成和评估来支持这一过程。然而，仅靠这些迭代并不能确保累积进展，也无法确定接下来应探索的方向。昂贵的评估进一步限制了探索的范围。神经架构发现将这些挑战汇集在一起，将开放式设计与资源密集型实验相结合。我们提出了EvoTreeNAD，一种谱系引导的进化算法，无需提供种子或手工指定的搜索空间即可构建可训练的架构。从空根节点开始，它生长出一个持久的谱系，其中每个新节点代表一个完整的架构。由每个节点及其后代计算出的前百分位值指导谱系选择。利用选定的设计历史，一个想法智能体提出变体，代码智能体实现该变体。每个评估过的变体成为子节点，扩展谱系的同时为后续谱系选择提供证据。我们的理论分析确立了随着谱系增长，平稳变异机制的存在。在指定的变异假设下，持续的前百分位家族值量化了在这些机制中生成高奖励架构的概率。EvoTreeNAD发现的架构优于所比较的NAS和NAD基线，在CIFAR-10/100上实现了$2.05{\pm}0.06\\%$和$15.09{\pm}0.22\\%$的测试错误率。在所有六个MedMNIST-v2任务上，发现的架构超越了最强的列出的基线。一项受控的CIFAR-10研究进一步表明，EvoTreeNAD优于直接生成、最佳-$N$贪心延续和全家族均值路由。
英文摘要
AI-driven scientific discovery accelerates research by autonomously developing solutions and designs. Large language model (LLM) agents support this process through iterative generation and evaluation. Yet these iterations alone do not ensure cumulative progress or establish which directions to pursue next. Costly evaluation further constrains the scope of exploration. Neural architecture discovery brings these challenges together, coupling open-ended design with resource-intensive experimentation. We introduce EvoTreeNAD, a genealogy-guided evolutionary algorithm that constructs trainable architectures without a supplied seed or a hand-specified search space. Starting from an empty root, it grows a persistent genealogy in which each new node represents a complete architecture. Top-percentile values computed from each node and its descendants guide lineage selection. Using the selected design history, an Idea Agent proposes a variant and a Code Agent implements it. Each evaluated variant becomes a child node, expanding the genealogy while providing evidence for subsequent lineage selection. Our theoretical analysis establishes the existence of stationary variation regimes as the genealogy grows. Under specified variation assumptions, sustained top-percentile family values quantify the probability of generating high-reward architectures in these regimes. EvoTreeNAD discovers architectures that outperform the compared NAS and NAD baselines, achieving CIFAR-10/100 test errors of $2.05{\pm}0.06\%$ and $15.09{\pm}0.22\%$. On all six MedMNIST-v2 tasks, the discovered architectures surpass the strongest listed baselines. A controlled CIFAR-10 study further shows that EvoTreeNAD outperforms direct generation, best-of-$N$ greedy continuation, and full-family-mean routing.
Comments31 pages, 5 figures, including appendices
