Physics-Audited Agentic Discovery in Scientific Machine Learning

📄 arXiv: 2607.07379v1 📥 PDF

作者: Diab W. Abueidda, Bilal Ahmed, Panos Pantidis, Mostafa E. Mobasher

分类: cs.AI, cs.LG

发布日期: 2026-07-08


💡 一句话要点

提出物理审核的代理科学机器学习以解决模型验证问题

🎯 匹配领域: 支柱九:具身大模型 (Embodied Foundation Models)

关键词: 代理科学机器学习 物理审核 模型验证 数值模拟 固体力学

📋 核心要点

  1. 现有的代理科学机器学习方法在模型选择时仅依赖误差评分,未能确保物理条件的满足。
  2. 本文提出PA-SciML工作流程,通过物理审核确保模型输出符合物理要求,增强了模型的可靠性。
  3. 实验结果表明,所选替代模型在验证误差上优于传统方法,同时通过了更严格的物理检查。

📝 摘要(中文)

在代理科学机器学习(SciML)中,大型语言模型(LLM)代理能够发现替代模型并通过自动评分选择模型。然而,低误差并不意味着预测结果满足力学中的物理条件。本文提出了物理审核的代理SciML(PA-SciML),一种以验证为先的工作流程,确保在搜索之前固定评分评估器,推导可审查的机器可检查物理要求,并检查每个训练候选模型的输出。该方法在计算固体力学的数值示例中表现出色,所选替代模型在验证误差上优于仅基于误差的基线,同时满足线性弹性检查。

🔬 方法详解

问题定义:本文旨在解决现有代理科学机器学习方法在模型验证中的不足,特别是仅依赖误差评分而忽视物理条件的满足。

核心思路:PA-SciML工作流程通过在搜索之前固定评分评估器,并推导可审查的物理要求,确保每个候选模型的输出符合物理条件。

技术框架:该工作流程包括以下主要模块:固定评分评估器、物理要求推导、候选模型输出检查,以及高违例案例的搜索。

关键创新:最重要的创新在于引入了物理审核机制,使得每个替代模型在被报告为验证通过之前,必须满足特定的物理检查,这与传统方法的误差评分机制本质不同。

关键设计:在模型训练前,工作流程会添加数值探针,并逐一测试建模变更,以记录哪些独立的编辑与评分提升相关联。

🖼️ 关键图片

fig_0
fig_1
fig_2

📊 实验亮点

实验结果显示,在静态弹性测试中,所选替代模型的验证误差低于仅基于误差的基线,同时通过了常见的线性弹性检查。在瞬态弹性动力学测试中,传统基线模型未能通过因果性检查,而所选替代模型则成功通过。

🎯 应用场景

该研究的潜在应用领域包括工程设计、材料科学和结构分析等领域,能够提高模型的可靠性和准确性,确保在实际应用中满足物理条件,从而推动科学计算和机器学习的结合。

📄 摘要(原文)

In agentic scientific machine learning (SciML), large language model (LLM) agents can discover surrogate models and select one by an automated score, typically an error metric. A low error, however, does not establish that the predicted fields satisfy the physics that matter for mechanics, such as boundary conditions, superposition, stiffness scaling, or causality. We introduce Physics-Audited Agentic SciML (PA-SciML), a verification-first workflow for agentic SciML discovery. The workflow fixes a scoring evaluator before search, derives reviewable machine-checkable physics requirements, checks each trained candidate on its outputs, and separately searches prescribed input ranges or measured load-history spans for high-violation cases without reference solution fields. A surrogate is reported as verified only under the stated checks. When enabled, the workflow also adds advisory numerical probes before training and tests one modeling change at a time to record which isolated edits are associated with score gains before reuse. In the reported computational-solid-mechanics numerical examples, the static elasticity run selects a surrogate with lower validation error than the error-only baseline while both selected models pass the common linear-elastic checks. In the transient elastodynamics run, an error-only baseline with similar mean error fails a stricter causality check by responding to future parts of the loading history, while the selected surrogate passes the stated checks. The main distinction is per-candidate physics evidence on predicted fields, not a richer aggregate score.