自动搜索分子模型架构反而不如调学习率，且发现的结构可跨领域通用。
What an Autonomous Agent Discovers About Molecular Transformer Design: Does It Transfer?
- 用智能体在三种序列上自动搜索模型架构
- 调学习率比全搜索效果更好，跨域性能下降不足1%
- 适合想省时省力的药物设计团队使用
针对类药物分子和蛋白质的深度学习模型普遍复用自然语言处理中的Transformer架构，但其对分子序列是否真正受益于不同设计尚未系统验证。我们通过智能体在三种序列类型（SMILES、蛋白质、英文文本）上进行自主架构搜索，单块GPU运行3,106次实验。结果显示：在SMILES任务中，架构搜索反而不利，仅调整学习率和调度策略就优于完整搜索（p=0.001）；在自然语言任务中，架构改进贡献了81%的性能提升（p=0.009）；蛋白质介于两者之间。令人意外的是，尽管智能体为各领域发现了不同架构（p=0.004），但所有创新在跨域迁移时性能损失均小于1%，表明差异源于搜索路径依赖而非生物本质需求。我们发布了决策框架与开源工具包，帮助分子建模团队在自动架构搜索与简单超参数调优间做出选择。
原文摘要 · Abstract (English)
Deep learning models for drug-like molecules and proteins overwhelmingly reuse transformer architectures designed for natural language, yet whether molecular sequences benefit from different designs has not been systematically tested. We deploy autonomous architecture search via an agent across three sequence types (SMILES, protein, and English text as control), running 3,106 experiments on a single GPU. For SMILES, architecture search is counterproductive: tuning learning rates and schedules alone outperforms the full search (p = 0.001). For natural language, architecture changes drive 81% of improvement (p = 0.009). Proteins fall between the two. Surprisingly, although the agent discovers distinct architectures per domain (p = 0.004), every innovation transfers across all three domains with <1% degradation, indicating that the differences reflect search-path dependence rather than fundamental biological requirements. We release a decision framework and open-source toolkit for molecular modeling teams to choose between autonomous architecture search and simple hyperparameter tuning.
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。