New framework decouples physics from discrepancy learning for system identification

Researchers have introduced Orthogonal Discrepancy Kernels (ODKs), a novel semi-parametric framework designed for nonlinear system identification. This approach effectively separates discrepancy functions from physics-based components, enabling more interpretable models even with incomplete physical data. The framework utilizes orthogonal Gaussian process regression to balance sparse parameter selection with discrepancy learning. AI

RANK_REASON The cluster contains a research paper published on arXiv detailing a new framework for system identification. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv stat.ML →

• arXiv
• Orthogonal Discrepancy Kernels
• Swapnil Manna

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →