Parameter values for seismic processing steps are often chosen on a regular grid of samples and interpolated. Active learning instead attempts to optimally select the samples on which parameter values are chosen. For parameters that do not vary smoothly, this often reduces the number of samples that need to be labelled in order to achieve a desired accuracy on the whole dataset. In regression tasks this is typically achieved using a query by committee strategy that selects the samples on which a committee of models is most uncertain. I implement such a strategy for the first break picking task, where the parameters to be chosen are the centre and width of the picking window for each trace. For the committee members I use the centre of the picking window and three popular picking algorithms. Applying this to a real dataset, and with samples corresponding to shot gathers, the active learning approach primarily selects gathers near a jump in the first breaks, and achieves similar levels of accuracy on the whole dataset with about half the number samples picked as when the samples are randomly selected.
Seismic First Break Picking in a Higher Dimension Using Deep Graph Learning
School of Mathematics and Statistics, Xi’an Jiaotong University（西安交通大学数学与统计学院）；China National Petroleum Corp Bureau of Geophysical Prospecting Inc, Geophysical Technology Research Center（中国石油集团地球物理勘探局有限公司地球物理技术研究中心）
GreenPhase: A Green Learning Approach for Earthquake Phase Picking
Ming Hsieh Department of Electrical and Computer Engineering, University of Southern California（美国南加州大学明希斯电气与计算机工程系）；Center for Earth and Environmental Studies, National Central University（中华大学地球与环境研究所）；Department of Computer Science & Information Engineering, National Central University（中华大学计算机科学与信息工程系）
SeisT: A foundational deep learning model for earthquake monitoring tasks
School of Computer Science and Technology, China University of Mining and Technology（中国矿业大学计算机科学与技术学院）；School of Mines, China University of Mining and Technology（中国矿业大学矿业工程学院）；State Key Laboratory of Coal Resources and Safe Mining, China University of Mining and Technology（中国矿业大学煤炭资源与安全开采国家重点实验室）
StorSeismic: A new paradigm in deep learning for seismic processing
King Abdullah University of Science and Technology（阿卜杜拉国王科技大学）
地震数据上的机器学习任务通常是顺序和单独训练的，即使它们利用数据的相同特征（即几何特征）。我们提出StorSeismic，作为地震数据处理的框架，它包括神经网络预训练和微调程序。具体而言，我们利用神经网络作为预处理模型来存储特定数据集的地震数据特征以用于任何下游任务。在预训练之后，所得到的模型可以稍后通过微调过程来使用，以使用有限的额外训练来执行任务。BERT（Bidirectional Encoder Representations from Transformer）是一种Transformer模型的形式，经常用于自然语言处理（NLP），最近用于视觉任务，为该框架提供了最佳平台。BERT的注意力机制，在这里应用于炮集内的一系列道，能够捕获和存储地震数据的关键几何特征。在自我监督步骤中，我们根据现场数据以及综合生成的数据对StorSeismic进行预训练。然后，我们使用标记的合成数据以监督的方式微调预训练的网络，以执行各种地震处理任务，如去噪、速度估计、初至拾取和动校正。最后，使用微调后的模型对现场数据进行推理，得到了满意的结果。
