State-observation augmented diffusion model for nonlinear assimilation with unknown dynamics
This work addresses data assimilation problems for fields like weather forecasting or environmental monitoring, but it appears incremental as it builds on prior score-based approaches.
The paper tackles the challenge of data assimilation in highly nonlinear physical and observational models by proposing the State-Observation Augmented Diffusion (SOAD) model, which theoretically matches the true posterior distribution and shows improved performance over existing data-driven methods.
Data assimilation has become a key technique for combining physical models with observational data to estimate state variables. However, classical assimilation algorithms often struggle with the high nonlinearity present in both physical and observational models. To address this challenge, a novel generative model, termed the State-Observation Augmented Diffusion (SOAD) model is proposed for data-driven assimilation. The marginal posterior associated with SOAD has been derived and then proved to match the true posterior distribution under mild assumptions, suggesting its theoretical advantages over previous score-based approaches. Experimental results also indicate that SOAD may offer improved performance compared to existing data-driven methods.