用条件高斯方法改进数据同化，性能远超深度学习模型。
A competitive baseline for deep learning enhanced data assimilation using conditional Gaussian ensemble Kalman filtering
- 用条件高斯公式重构卡尔曼增益，突破传统线性假设。
- 在洛伦兹96系统中，新方法精度显著优于深度学习粒子滤波器。
- 可处理高度非高斯噪声，适合复杂多尺度系统同化任务。
集合卡尔曼滤波（EnKF）是数据同化中的常用技术，但其原始框架在非线性扰动下定义不清。本文研究了两种非线性扩展——条件高斯集合卡尔曼滤波（CG-EnKF）与正态得分集合卡尔曼滤波（NS-EnKF），通过用条件高斯更新公式替代传统公式来避免线性假设。将这两种方法与基于深度学习的评分滤波器（Score Filter, SF）进行比较，后者依赖昂贵的得分扩散模型并需强扰动算子假设。在高维多尺度数据同化基准问题（洛伦兹96系统）中，CG-EnKF和NS-EnKF表现大幅优于SF。分析还表明，两者能有效处理高度非高斯加性噪声，其中NS-EnKF通常优于CG-EnKF。
原文摘要 · Abstract (English)
Ensemble Kalman Filtering (EnKF) is a popular technique for data assimilation, with far ranging applications. However, the vanilla EnKF framework is not well-defined when perturbations are nonlinear. We study two non-linear extensions of the vanilla EnKF - dubbed the conditional-Gaussian EnKF (CG-EnKF) and the normal score EnKF (NS-EnKF) - which sidestep assumptions of linearity by constructing the Kalman gain matrix with the `conditional Gaussian' update formula in place of the traditional one. We then compare these models against a state-of-the-art deep learning based particle filter called the score filter (SF). This model uses an expensive score diffusion model for estimating densities and also requires a strong assumption on the perturbation operator for validity. In our comparison, we find that CG-EnKF and NS-EnKF dramatically outperform SF for a canonical problem in high-dimensional multiscale data assimilation given by the Lorenz-96 system. Our analysis also demonstrates that the CG-EnKF and NS-EnKF can handle highly non-Gaussian additive noise perturbations, with the latter typically outperforming the former.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。