New adaptive surrogate modeling tackles high-dimensional spatio-temporal data

Researchers have developed a new adaptive surrogate modeling method designed to handle problems with extremely high-dimensional spatio-temporal outputs. This approach first reduces the dimensionality of the output data into a lower-dimensional latent space before constructing a surrogate model. The method then adaptively identifies new training points to improve the model's accuracy with minimal calls to the expensive physics-based model, demonstrating its effectiveness through a thermo-mechanical analysis of a gas turbine engine blade. AI

IMPACT This method could improve the efficiency of complex simulations in engineering and scientific fields by reducing computational costs.

RANK_REASON The cluster contains a research paper published on arXiv detailing a new methodology. [lever_c_demoted from research: ic=1 ai=0.7]

Read on arXiv stat.ML →

• arXiv
• Berkcan Kapusuzoğlu
• Computational Engineering, Finance, and Science

• paper
• other

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