Researchers have developed a new active learning framework designed to improve data-driven reduced-order models (ROMs) for parametric dynamical systems. This framework uses Bayesian operator inference, framed as Bayesian linear regression, to create probabilistic ROMs. By analyzing prediction uncertainties, the system adaptively selects new training parameters to enhance ROM stability and accuracy, outperforming random sampling in numerical experiments with partial differential equations. AI
IMPACT This research could lead to more accurate and stable digital twins and simulations by improving the efficiency of training reduced-order models.
RANK_REASON The item is an academic paper detailing a new methodology for scientific machine learning. [lever_c_demoted from research: ic=1 ai=1.0]
- arXiv
- Bayesian linear regression
- digital twin
- Operator Inference
- parametric dynamical systems
- partial differential equations
- Scientific Machine Learning
- Shane McQuarrie
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
