New generative learning framework enhances seismic performance assessment

Researchers have developed a new manifold-based generative framework that uses Diffusion Maps, score-based Diffusion Modeling, and Geometric Harmonics to learn the joint distribution of structural parameters and seismic-response trajectories. This approach aims to reduce the computational cost of seismic performance assessment and improve uncertainty quantification. Applied to a nine-story steel moment-resisting frame, the framework can predict forward probabilistic outcomes and perform inverse inference to identify which structural properties are most influential and identifiable from observed seismic performance. AI

IMPACT This research demonstrates how generative learning can improve the efficiency and accuracy of seismic performance assessment, potentially leading to more robust structural designs.

RANK_REASON Academic paper detailing a new methodology for seismic performance assessment. [lever_c_demoted from research: ic=1 ai=0.7]

Read on arXiv cs.LG →

• alphaXiv
• arXiv
• CatalyzeX
• CORE Recommender
• DagsHub
• Dimitris Giovanis
• Geometric Harmonics
• Gotit.pub
• Hugging Face
• Influence Flower
• nine-story steel moment-resisting frame
• ScienceCast
• score-based Diffusion Modeling

• paper
• infra

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