Bayesian Optimisation for Structures: Challenges and Recommendations

Add to your list(s) Download to your calendar using vCal

• Thomas Archbold, University of Cambridge
• Friday 29 October 2021, 15:00-16:00
• Zoom (email structures-admin@eng.cam.ac.uk for link).

If you have a question about this talk, please contact Mishael Nuh.

Structural optimisation is an essential tool for reducing greenhouse gas emissions from the building and transportation sectors. Many finite element models used for simulation are computationally expensive and are called “black- box” because of their lack of derivative information. This limits the applicability of many gradient-based methods. Bayesian optimisation is a “data-driven” approach used to efficiently optimise “black-box” objective functions that are computationally expensive to evaluate. Structural optimisation problems can be high-dimensional and scaling Bayesian optimisation to such settings remains a barrier to its wide-spread use in practice. This project aims to explore how Bayesian optimisation can be used to solve high- dimensional structural optimisation problems.

This talk is part of the Engineering Department Structures Research Seminars series.

This talk is included in these lists:

• All Talks (aka the CURE list)
• Cambridge University Engineering Department Talks
• Cambridge talks
• Centre for Smart Infrastructure & Construction
• Civil Engineering Talks
• Computational Continuum Mechanics Group Seminars
• Engineering Department Structures Research Seminars
• Featured lists
• Interested Talks
• School of Technology
• Trust & Technology Initiative - interesting events
• Zoom (email structures-admin@eng.cam.ac.uk for link)
• bld31

Note that ex-directory lists are not shown.