Get Started
Home
Topics
Search
Library
Research questionHow should training parameters be selected to keep data-driven reduced-order models stable and accurate across a parameter domain?Reduced-order models depend strongly on which parameter settings supply their limited training data. Poor coverage can produce models that are inaccurate or unstable for parameter values outside the sampled set.
AI
Machine Learning
Research Paper
Statistical Machine Learning
Technology
Latest papersRecent research connected to this question, newest first.Active learning for data-driven reduced models of parametric differential systems with Bayesian operator inferenceThe source concerns operator-inference reduced models and uses Bayesian prediction uncertainty to guide sampling. Evidence comes from numerical experiments on nonlinear parametric partial differential equation systems, compared with random parameter sampling under the same computational budget.research paper · Sep 3, 2026
Related questions
How can learned chaotic systems preserve long-term statistics and stability beyond short-term trajectory matching?When do tied or untied attention parameterizations enable weak recovery under stochastic training?How can reliable foundation-model scaling laws be constructed without training every configuration?How should momentum and batch size be tuned to preserve stability and data efficiency in one-pass training?