Latent assimilation with implicit neural representations for unknown dynamics
This work addresses data assimilation problems in applications where efficiency and accuracy are critical, representing an incremental improvement over existing methods.
The paper tackles the challenges of high computational costs and incomplete understanding in data assimilation by proposing LAINR, a framework using Spherical Implicit Neural Representations and a data-driven uncertainty estimator, which shows advantages in accuracy and efficiency over AutoEncoder-based methods.
Data assimilation is crucial in a wide range of applications, but it often faces challenges such as high computational costs due to data dimensionality and incomplete understanding of underlying mechanisms. To address these challenges, this study presents a novel assimilation framework, termed Latent Assimilation with Implicit Neural Representations (LAINR). By introducing Spherical Implicit Neural Representations (SINR) along with a data-driven uncertainty estimator of the trained neural networks, LAINR enhances efficiency in assimilation process. Experimental results indicate that LAINR holds certain advantage over existing methods based on AutoEncoders, both in terms of accuracy and efficiency.