Attention-based Convolutional Autoencoders for 3D-Variational Data Assimilation

Julian Mack, Rossella Arcucci, Miguel Molina-Solana +1

We propose a new 'Bi-Reduced Space' approach to solving 3D Variational Data Assimilation using Convolutional Autoencoders. We prove that our approach has the same solution as previ…