arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2605.11280gr-qcastro-ph.HEcs.AI
收藏

通过代理AI发现可解释的替代模型:应用于引力波

Discovery of Interpretable Surrogates via Agentic AI: Application to Gravitational Waves

Kavli Institute for Theoretical Physics, University of California Santa Barbara(加州大学圣芭芭拉分校凯文利理论物理研究所);Center for Gravitational Physics, University of Texas at Austin(德克萨斯大学奥斯汀分校重力物理中心);Department of Physics, University of California at Santa Barbara(加州大学圣芭芭拉分校物理系);International Centre for Theoretical Sciences, Tata Institute of Fundamental Research(塔塔基础研究机构国际理论科学研究中心);School of Natural Sciences, Institute for Advanced Study(高级研究院自然科学学院);Chennai Mathematical Institute(钦奈数学研究所);Kavli Institute for Cosmological Physics, The University of Chicago(芝加哥大学凯文利宇宙物理研究所);Department of Particle Physics & Astrophysics, Weizmann Institute of Science(魏茨曼科学研究所粒子物理与天体物理系)

Tousif Islam, Digvijay Wadekar, Tejaswi Venumadhav, Matias Zaldarriaga, Ajit Kumar Mehta, Javier Roulet, Barak Zackay

AI总结:

本文提出GWAgent,一种基于大语言模型的流程,直接从模拟数据构建可解释的分析替代模型。通过物理指导的域假设提升模型精度,实现高精度、快速且可解释的引力波模拟替代。

摘要:

快速的替代模型对于昂贵的模拟现在是各学科中不可或缺的,但它们通常作为黑箱操作。我们提出了GWAgent,一种基于大语言模型(LLM)的工作流程,直接从模拟数据构建可解释的分析替代模型。替代建模适合代理工作流程,因为候选模型可以在每次迭代中通过与真实模拟的定量验证进行验证。作为演示,我们构建了一个从偏心二体黑洞合并产生的引力波形的替代模型。我们展示了给代理提供物理指导的域假设显著提高了输出模型的准确性。所得到的分析替代模型达到了中位数Advanced LIGO不匹配度为6.9×10⁻⁴,同时在波形评估中实现了约8.4倍的速度提升,超过了符号回归和传统机器学习基线。除了产生准确的模型外,该工作流程还从学习的表示中识别出紧凑的物理结构。作为天体物理应用,我们使用GWAgent分析GW200129的偏心率,并推断出e₂₀Hz=0.099±0.063⁻⁰.044。这些结果表明,受验证约束的代理工作流程能够为科学模拟和推断产生准确、快速且可解释的替代模型。

英文摘要:

Fast surrogate models for expensive simulations are now essential across the sciences, yet they typically operate as black boxes. We present \texttt{GWAgent}, a large language model (LLM)-based workflow that constructs interpretable analytic surrogates directly from simulation data. Surrogate modeling is well suited to agentic workflows because candidate models can be quantitatively validated against ground-truth simulations at each iteration. As a demonstration, we build a surrogate for gravitational waveforms from eccentric binary black hole mergers. We show that providing the agent with a physics-informed domain ansatz substantially improves output model accuracy. The resulting analytic surrogate attains a median Advanced LIGO mismatch of $6.9\times10^{-4}$ together with an $\sim 8.4\times$ speedup in waveform evaluation, surpassing both symbolic regression and conventional machine learning baselines. Beyond producing an accurate model, the workflow identifies compact physical structure from the learned representation. As an astrophysical application, we use \texttt{GWAgent} to analyze the eccentricity of GW200129 and infer $e_{20\mathrm{Hz}}=0.099^{+0.063}_{-0.044}$. These results show that validation-constrained agentic workflows can produce accurate, fast, and interpretable surrogates for scientific simulations and inference.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑