Challenges in Bayesian inference and reliability with large numbers of uncertain parameters

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

• Jan Koune, TU Delft
• Friday 17 November 2023, 16:00-17:00
• JDB Seminar Room, CUED.

If you have a question about this talk, please contact div-c.

Efficient sampling methods play a central role in Bayesian inference, reliability analysis and machine learning. Practical applications in structural and geotechnical engineering often deal with high-dimensional parameter spaces and computationally expensive physics-based models, rendering many sampling-based approaches infeasible.

In this talk we initially examine these challenges through the case studies of a steel girder road bridge and a sheet pile quay wall. Motivated by recent work on machine learning approaches for nested sampling, we explore the connection between nested sampling and importance sampling, subset simulation and normalizing flows, in an effort to derive an optimization-based approach for inference and reliability estimation that takes advantage of the key ideas behind nested sampling.

This talk is part of the Engineering - Dynamics and Vibration Tea Time Talks 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 - Dynamics and Vibration Tea Time Talks
• Engineering - Mechanics, Materials and Design (Div C) - talks and events
• Featured lists
• Interested Talks
• JDB Seminar Room, CUED
• School of Technology
• Trust & Technology Initiative - interesting events
• bld31

Note that ex-directory lists are not shown.