Generative models for simulation based filtering: Formulations and Empirical Comparisons
A unified formulation and a controlled numerical comparison of generative-model approaches to the nonlinear filtering problem are presented, indicating that every generative filter resolves multimodal posteriors that the EnKF and SIR do not, and that no single generative framework dominates.
Mohammad Al-Jarrah, Wei Deng, Bamdad Hosseini et al.
· 0 citations