Search Results for author:
Found 2 papers, 0 papers with code
Adjoint-Matching Neural Network Surrogates for Fast 4D-Var Data Assimilation
Surrogates constructed using adjoint information demonstrate superior performance on the 4D-Var data assimilation problem compared to a standard neural network surrogate that uses only forward dynamics information.
Investigation of Nonlinear Model Order Reduction of the Quasigeostrophic Equations through a Physics-Informed Convolutional Autoencoder
Reduced order modeling (ROM) is a field of techniques that approximates complex physics-based models of real-world processes by inexpensive surrogates that capture important dynamical characteristics with a smaller number of degrees of freedom.
Cannot find the paper you are looking for? You can
Submit a new open access paper.
