Machine learning-based seismic processing models are typically trained separately to perform seismic processing tasks (SPTs) and, as a result, require plenty of high-quality training data. However, preparing training data sets is not trivial, especially for supervised learning (SL). Despite the variability in seismic data across different types and regions, some general characteristics are shared, such as their sinusoidal nature and geometric texture. To learn the shared features and thus, quickly adapt to various SPTs, we develop a unified paradigm for neural network-based seismic processing, called Meta-Processing, that uses limited training data for meta learning a common network initialization, which offers universal adaptability features. The proposed Meta-Processing framework consists of two stages: meta-training and meta-testing. In the former, each SPT is treated as a separate task and the training dataset is divided into support and query sets. Unlike conventional SL methods, here, the neural network (NN) parameters are updated by a bilevel gradient descent from the support set to the query set, iterating through all tasks. In the meta-testing stage, we also utilize limited data to fine-tune the optimized NN parameters in an SL fashion to conduct various SPTs, such as denoising, interpolation, ground-roll attenuation, image enhancement, and velocity estimation, aiming to converge quickly to ideal performance. Extensive numerical experiments are conducted to assess the effectiveness of Meta-Processing on both synthetic and real-world data. The findings reveal that our approach leads to a substantial improvement in the convergence speed and predictive performance of the NN.

基于机器学习的地震处理模型通常单独训练来执行地震处理任务 (SPT)，因此需要大量高质量的训练数据。然而，准备训练数据集并不简单，特别是对于监督学习（SL）。尽管不同类型和地区的地震数据存在差异，但一些通用特征是相同的，例如正弦曲线性质和几何纹理。为了学习共享特征，从而快速适应各种 SPT，我们开发了一种基于神经网络的地震处理的统一范例，称为元处理，它使用有限的训练数据来元学习公共网络初始化，提供通用的适应性特征。所提出的元处理框架由两个阶段组成：元训练和元测试。在前者中，每个 SPT 被视为一个单独的任务，训练数据集被分为支持集和查询集。与传统的 SL 方法不同，这里的神经网络 (NN) 参数通过从支持集到查询集的双层梯度下降进行更新，迭代所有任务。在元测试阶段，我们还利用有限的数据以SL方式微调优化的神经网络参数，以进行各种SPT，例如去噪、插值、地滚衰减、图像增强和速度估计，旨在收敛快速达到理想的性能。进行了大量的数值实验来评估元处理对合成数据和真实数据的有效性。研究结果表明，我们的方法可以显着提高神经网络的收敛速度和预测性能。