Machine Learning Techniques to Construct Patched Analog Ensembles for Data Assimilation
| Autor: | L. Minah Yang, Ian Grooms |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
FOS: Computer and information sciences
Computer Science - Machine Learning Physics and Astronomy (miscellaneous) Computer science FOS: Physical sciences 010103 numerical & computational mathematics Machine learning computer.software_genre Statistics - Computation 01 natural sciences Machine Learning (cs.LG) Data assimilation Square root Component (UML) 0101 mathematics Computation (stat.CO) Numerical Analysis business.industry Applied Mathematics Filter (signal processing) Construct (python library) Computational Physics (physics.comp-ph) Computer Science Applications 010101 applied mathematics Computational Mathematics Generative model Modeling and Simulation 68T07 62M45 93E11 Scalability Artificial intelligence business Physics - Computational Physics computer Interpolation |
| DOI: | 10.48550/arxiv.2103.00318 |
| Popis: | Using generative models from the machine learning literature to create artificial ensemble members for use within data assimilation schemes has been introduced in [Grooms QJRMS, 2020] as constructed analog ensemble optimal interpolation (cAnEnOI). Specifically, we study general and variational autoencoders for the machine learning component of this method, and combine the ideas of constructed analogs and ensemble optimal interpolation in the data assimilation piece. To extend the scalability of cAnEnOI for use in data assimilation on complex dynamical models, we propose using patching schemes to divide the global spatial domain into digestible chunks. Using patches makes training the generative models possible and has the added benefit of being able to exploit parallelism during the generative step. Testing this new algorithm on a 1D toy model, we find that larger patch sizes make it harder to train an accurate generative model (i.e. a model whose reconstruction error is small), while conversely the data assimilation performance improves at larger patch sizes. There is thus a sweet spot where the patch size is large enough to enable good data assimilation performance, but not so large that it becomes difficult to train an accurate generative model. In our tests the new patched cAnEnOI method outperforms the original (unpatched) cAnEnOI, as well as the ensemble square root filter results from [Grooms QJRMS, 2020]. Comment: 21 pages, 10 figures |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|