标题： 机器学习初始化加速斯托克斯轮廓反演 显示英文标题

标题： Machine learning initialization to accelerate Stokes profile inversions

摘要： 在这项工作中，我们讨论了使用卷积神经网络（CNNs）作为有利初始化斯托克斯轮廓反演的工具。为了展示CNNs的有用性，本文集中于非局部热平衡（LTE）斯托克斯轮廓的反演。我们使用搭载于“日出”航天器上的光谱偏振仪观测数据作为测试基准。首先，我们使用给定的初始大气模型，仔细分析数据并使用SIR反演代码。该代码提供了一组能够再现观测的大气模型。然后这些模型被用于训练CNN。随后，相同的数据再次用SIR反演，但这次使用训练好的CNN来提供SIR的初始猜测大气模型。当用于计算初始猜测模型大气时，CNN显著减少了反演循环次数，使LTE反演的计算时间减少到原来的二分之一到四分之一。单独使用CNN的速度比辅助反演快得多，但后者更稳健和准确。讨论了机器学习技术在估算光谱线反演的最佳初始大气模型方面的优势和局限性。最后，我们描述了一个用于SIR和DeSIRe代码的Python包装器，以方便并行反演的设置。辅助反演可以加速反演过程，但反演结果的效率和准确性强烈依赖于太阳场景和用于CNN训练的数据。这种方法（辅助反演）不会消除对个别事件进行最细致分析的需求，但将为太阳物理学家提供更好的机会来采样大量反演数据，这无疑会拓宽物理发现的空间。 显示英文摘要

摘要： In this work, we discuss the application of convolutional neural networks (CNNs) as a tool to advantageously initialize Stokes profile inversions. To demonstrate the usefulness of CNNs, we concentrate in this paper on the inversion of LTE Stokes profiles. We use observations taken with the spectropolarimeter onboard the Hinode spacecraft as a test benchmark. First, we carefully analyze the data with the SIR inversion code using a given initial atmospheric model. The code provides a set of atmospheric models that reproduce the observations. These models are then used to train a CNN. Afterwards, the same data are again inverted with SIR but using the trained CNN to provide the initial guess atmospheric models for SIR. The CNNs allow us to significantly reduce the number of inversion cycles when used to compute initial guess model atmospheres, decreasing the computational time for LTE inversions by a factor of two to four. CNN's alone are much faster than assisted inversions, but the latter are more robust and accurate. The advantages and limitations of machine learning techniques for estimating optimum initial atmospheric models for spectral line inversions are discussed. Finally, we describe a python wrapper for the SIR and DeSIRe codes that allows for the easy setup of parallel inversions. The assisted inversions can speed up the inversion process, but the efficiency and accuracy of the inversion results depend strongly on the solar scene and the data used for the CNN training. This method (assisted inversions) will not obviate the need for analyzing individual events with the utmost care but will provide solar scientists with a much better opportunity to sample large amounts of inverted data, which will undoubtedly broaden the physical discovery space.