Using the SEKF to Transfer NN Models of Dynamical Systems with Limited Data

2026-03-02

AI Summary

This work addresses the high data acquisition cost often encountered in data-driven modeling of dynamical systems by proposing an efficient fine-tuning approach based on Subset Extended Kalman Filtering (SEKF). The method enables successful transfer of a pretrained neural network model to a new dynamical system using only 1% of the original training data. By integrating Bayesian estimation with dynamical system modeling, SEKF substantially reduces both data requirements and computational overhead while enhancing model generalization. Experimental validation on a damped spring-mass system and a continuous stirred-tank reactor demonstrates that fine-tuning only a small subset of parameters suffices to accurately capture the target dynamics, thereby confirming the efficacy and practicality of the proposed approach.

Technology Category

• Intelligent Robots: State Estimation
• Machine Learning: Learning with Manifolds
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

Data-driven models of dynamical systems require extensive amounts of training data. For many practical applications, gathering sufficient data is not feasible due to cost or safety concerns. This work uses the Subset Extended Kalman Filter (SEKF) to adapt pre-trained neural network models to new, similar systems with limited data available. Experimental validation across damped spring and continuous stirred-tank reactor systems demonstrates that small parameter perturbations to the initial model capture target system dynamics while requiring as little as 1% of original training data. In addition, finetuning requires less computational cost and reduces generalization error.

Problem

• data-driven modeling
• dynamical systems
• limited data
• neural network adaptation
• training data scarcity

Innovation

• Subset Extended Kalman Filter
• Neural Network Transfer
• Data-Efficient Adaptation
• Dynamical Systems Modeling
• Limited Data Fine-tuning