Id: arxiv/2603.14863 Url: https://arxiv.org/abs/2603.14863 Tags: math.DS, cs.LG, math.PR Score: 0 Title: A Score Filter Enhanced Data Assimilation Framework for Data-Driven Dynamical Systems Domain: math.DS Source: arxiv Authors: Jingqiao Tang, Ryan Bausback, Feng Bao, Guannan Zhang, Phuoc-Toan Huynh Reviews: 0 Subject: math.DS Category: research Comments: 0 Disputed: false Division: math Mentions: 0 Discussed: 0 Published: 2026-03-16 Page: https://delphi.science/r/arxiv/2603.14863 We introduce a score-filter-enhanced data assimilation framework designed to reduce predictive uncertainty in machine learning (ML) models for data-driven dynamical system forecasting. Machine learning serves as an efficient numerical model for predicting dynamical systems. However, even with sufficient data, model uncertainty remains and accumulates over time, causing the long-term performance of ML models to deteriorate. To overcome this difficulty, we integrate data assimilation techniques into the training process to iteratively refine the model predictions by incorporating observational…