arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2408.05425cs.LG
收藏

基于图Transformer网络的多步骤科学过程建模

Modeling Multi-Step Scientific Processes with Graph Transformer Networks

National Research Council(美国国家研究委员会);Air Force Research Laboratory(空军研究实验室);National Renewable Energy Laboratory(国家可再生能源实验室)

Amanda A. Volk, Robert W. Epps, Jeffrey G. Ethier, Luke A. Baldwin

AI总结:

本研究提出用图Transformer网络建模多步骤科学过程,经模拟与真实实验数据验证,其在多类场景下的实验结果预测表现优于线性模型,可为高维参数空间探索提供支持。

摘要:

本研究提出将图学习用于多步骤实验结果预测,可应用于材料科学、化学、生物学等多个实验研究领域。研究结合模拟数据与真实数据训练实验,将几何学习在回归任务中的可行性与一组线性模型进行基准对比。首先,研究设计了5个任意构造的多步骤代理函数,用以反映实验过程中常见的各类特征。在存在过程步骤间隐藏相互作用、具备序列依赖特征的场景中,图Transformer网络(graph transformer network)的表现优于所有受试线性模型;在与序列无关的场景中,其性能与线性模型相当。随后,研究针对算法引导的胶体原子层沉积(colloidal atomic layer deposition)的真实文献数据开展了同类对比。以完整反应序列作为训练数据时,在大多数训练集规模下,图神经网络(graph neural network)在预测3种光谱属性方面的表现均优于所有线性模型。进一步将图神经网络与科学过程的几何表示应用于实验结果预测,可实现算法驱动的高维参数空间导航,以及对更多动态系统的高效探索。

英文摘要:

This work presents the use of graph learning for the prediction of multi-step experimental outcomes for applications across experimental research, including material science, chemistry, and biology. The viability of geometric learning for regression tasks was benchmarked against a collection of linear models through a combination of simulated and real-world data training studies. First, a selection of five arbitrarily designed multi-step surrogate functions were developed to reflect various features commonly found within experimental processes. A graph transformer network outperformed all tested linear models in scenarios that featured hidden interactions between process steps and sequence dependent features, while retaining equivalent performance in sequence agnostic scenarios. Then, a similar comparison was applied to real-world literature data on algorithm guided colloidal atomic layer deposition. Using the complete reaction sequence as training data, the graph neural network outperformed all linear models in predicting the three spectral properties for most training set sizes. Further implementation of graph neural networks and geometric representation of scientific processes for the prediction of experiment outcomes could lead to algorithm driven navigation of higher dimension parameter spaces and efficient exploration of more dynamic systems.

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。

↑