Concrete Autoencoder for the Reconstruction of Sea Temperature Field from Sparse Measurements
This paper presents a new method for finding the optimal positions for sensors used to reconstruct geophysical fields from sparse measurements. The method is composed of two stages. In the first stage, we estimate the spatial variability of the physical field by approximating its information entropy...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-02-01
|
| Series: | Journal of Marine Science and Engineering |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2077-1312/11/2/404 |
| _version_ | 1797620025637470208 |
|---|---|
| author | Alexander A. Lobashev Nikita A. Turko Konstantin V. Ushakov Maxim N. Kaurkin Rashit A. Ibrayev |
| author_facet | Alexander A. Lobashev Nikita A. Turko Konstantin V. Ushakov Maxim N. Kaurkin Rashit A. Ibrayev |
| author_sort | Alexander A. Lobashev |
| collection | DOAJ |
| description | This paper presents a new method for finding the optimal positions for sensors used to reconstruct geophysical fields from sparse measurements. The method is composed of two stages. In the first stage, we estimate the spatial variability of the physical field by approximating its information entropy using the Conditional Pixel CNN network. In the second stage, the entropy is used to initialize the distribution of optimal sensor locations, which is then optimized using the Concrete Autoencoder architecture with the straight-through gradient estimator for the binary mask and with adversarial loss. This allows us to simultaneously minimize the number of sensors and maximize reconstruction accuracy. We apply our method to the global ocean under-surface temperature field and demonstrate its effectiveness on fields with up to a million grid cells. Additionally, we find that the information entropy field has a clear physical interpretation related to the mixing between cold and warm currents. |
| first_indexed | 2024-03-11T08:35:34Z |
| format | Article |
| id | doaj.art-4eac557bb355435cb26a18d9d26c93a5 |
| institution | Directory Open Access Journal |
| issn | 2077-1312 |
| language | English |
| last_indexed | 2024-03-11T08:35:34Z |
| publishDate | 2023-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Journal of Marine Science and Engineering |
| spelling | doaj.art-4eac557bb355435cb26a18d9d26c93a52023-11-16T21:28:41ZengMDPI AGJournal of Marine Science and Engineering2077-13122023-02-0111240410.3390/jmse11020404Concrete Autoencoder for the Reconstruction of Sea Temperature Field from Sparse MeasurementsAlexander A. Lobashev0Nikita A. Turko1Konstantin V. Ushakov2Maxim N. Kaurkin3Rashit A. Ibrayev4Skolkovo Institute of Science and Technology, 30 Bolshoy Boulevard, Moscow 121205, RussiaMoscow Institute of Physics and Technology, 9 Institutskiy Per., Dolgoprudny 141701, RussiaMoscow Institute of Physics and Technology, 9 Institutskiy Per., Dolgoprudny 141701, RussiaShirshov Institute of Oceanology, Russian Academy of Sciences, 36 Nakhimovsky Prospekt, Moscow 117997, RussiaMoscow Institute of Physics and Technology, 9 Institutskiy Per., Dolgoprudny 141701, RussiaThis paper presents a new method for finding the optimal positions for sensors used to reconstruct geophysical fields from sparse measurements. The method is composed of two stages. In the first stage, we estimate the spatial variability of the physical field by approximating its information entropy using the Conditional Pixel CNN network. In the second stage, the entropy is used to initialize the distribution of optimal sensor locations, which is then optimized using the Concrete Autoencoder architecture with the straight-through gradient estimator for the binary mask and with adversarial loss. This allows us to simultaneously minimize the number of sensors and maximize reconstruction accuracy. We apply our method to the global ocean under-surface temperature field and demonstrate its effectiveness on fields with up to a million grid cells. Additionally, we find that the information entropy field has a clear physical interpretation related to the mixing between cold and warm currents.https://www.mdpi.com/2077-1312/11/2/404concrete autoencoderoptimal sensor placementinformation entropyocean state reconstructionmutual informationsensitivity |
| spellingShingle | Alexander A. Lobashev Nikita A. Turko Konstantin V. Ushakov Maxim N. Kaurkin Rashit A. Ibrayev Concrete Autoencoder for the Reconstruction of Sea Temperature Field from Sparse Measurements Journal of Marine Science and Engineering concrete autoencoder optimal sensor placement information entropy ocean state reconstruction mutual information sensitivity |
| title | Concrete Autoencoder for the Reconstruction of Sea Temperature Field from Sparse Measurements |
| title_full | Concrete Autoencoder for the Reconstruction of Sea Temperature Field from Sparse Measurements |
| title_fullStr | Concrete Autoencoder for the Reconstruction of Sea Temperature Field from Sparse Measurements |
| title_full_unstemmed | Concrete Autoencoder for the Reconstruction of Sea Temperature Field from Sparse Measurements |
| title_short | Concrete Autoencoder for the Reconstruction of Sea Temperature Field from Sparse Measurements |
| title_sort | concrete autoencoder for the reconstruction of sea temperature field from sparse measurements |
| topic | concrete autoencoder optimal sensor placement information entropy ocean state reconstruction mutual information sensitivity |
| url | https://www.mdpi.com/2077-1312/11/2/404 |
| work_keys_str_mv | AT alexanderalobashev concreteautoencoderforthereconstructionofseatemperaturefieldfromsparsemeasurements AT nikitaaturko concreteautoencoderforthereconstructionofseatemperaturefieldfromsparsemeasurements AT konstantinvushakov concreteautoencoderforthereconstructionofseatemperaturefieldfromsparsemeasurements AT maximnkaurkin concreteautoencoderforthereconstructionofseatemperaturefieldfromsparsemeasurements AT rashitaibrayev concreteautoencoderforthereconstructionofseatemperaturefieldfromsparsemeasurements |