arXiv 2012.144052020-12-28astro-ph.SRcs.LG

基于形状特征工程的太阳耀斑预测

Shape-based Feature Engineering for Solar Flare Prediction

University of Colorado Boulder（科罗拉多大学博尔德分校）；The Santa Fe Institute（圣达菲研究所）

Varad Deshmukh, Thomas Berger, James Meiss, Elizabeth Bradley

备注：: To be published in Proceedings for Innovative Applications of Artificial Intelligence Conference 2021
发表信息：: AAAI Conference on Artificial Intelligence, 35(17), 2021, 15293-15300

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出利用计算拓扑和计算几何工具从太阳磁图图像中提取新颖的基于形状的特征，并在MLP神经网络中评估，发现其优于传统物理特征，两者结合可进一步提升太阳耀斑预测能力。

摘要：

太阳耀斑是由太阳表面活动区（ARs）的磁场爆发引起的。这些事件可能对人类活动产生重大影响，如果能有良好的预报提供足够的提前预警，其中许多影响是可以减轻的。迄今为止，基于机器学习的耀斑预测方法已将AR图像的基于物理的属性用作特征；最近，也有一些工作使用深度学习方法（如卷积神经网络）自动推导出的特征。我们描述了一组使用计算拓扑和计算几何工具从太阳磁图图像中提取的基于形状的新颖特征。我们在多层感知器（MLP）神经网络的背景下评估这些特征，并将其性能与传统的基于物理的属性进行比较。我们表明，这些抽象的基于形状的特征优于人类专家选择的特征，并且两种特征集的组合进一步提高了预测能力。

英文摘要：

Solar flares are caused by magnetic eruptions in active regions (ARs) on the surface of the sun. These events can have significant impacts on human activity, many of which can be mitigated with enough advance warning from good forecasts. To date, machine learning-based flare-prediction methods have employed physics-based attributes of the AR images as features; more recently, there has been some work that uses features deduced automatically by deep learning methods (such as convolutional neural networks). We describe a suite of novel shape-based features extracted from magnetogram images of the Sun using the tools of computational topology and computational geometry. We evaluate these features in the context of a multi-layer perceptron (MLP) neural network and compare their performance against the traditional physics-based attributes. We show that these abstract shape-based features outperform the features chosen by the human experts, and that a combination of the two feature sets improves the forecasting capability even further.

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。

AI总结：

摘要：

英文摘要：

相关论文