"Design for ABC and history matching with Gaussian processes"

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

• Dr Richard Wilkinson, Sheffield University
• Tuesday 01 March 2016, 14:30-15:30
• Large Seminar Room, 1st Floor, Institute of Public Health, University Forvie Site, Robinson Way, Cambridge.

If you have a question about this talk, please contact Alison Quenault.

Abstract: History matching and approximate Bayesian computation (ABC) are techniques for parameter estimation for complex models. In both cases, Gaussian processes can be used to accelerate inference by building an emulator/surrogate model either of the simulator itself, or of a summary such as the likelihood function. Design approaches for building emulators usually focus on how to build accurate global approximations. However, if our aim is calibration, then this can be wasteful. In this talk I will discuss the use of Gaussian processes, and describe design strategies that can be used for history matching and calibration, both in a deterministic and stochastic setting. Entropy based designs that minimize the uncertainty in the classification surface are particularly useful, and I will show how these can be efficiently computed using Bayesian optimisation.

This talk is part of the MRC Biostatistics Unit Seminars series.

This talk is included in these lists:

• All Talks (aka the CURE list)
• BHRU Annual Lecture 2015
• BHRU Annual Lecture 2016
• Biology
• Cambridge Centre for Data-Driven Discovery (C2D3)
• Cambridge Immunology
• Cambridge talks
• Cardiovascular Epidemiology Unit Special Seminars
• Chris Davis' list
• Department of Public Health and Primary Care
• Graduate-Seminars
• Interested Talks
• Large Seminar Room, 1st Floor, Institute of Public Health, University Forvie Site, Robinson Way, Cambridge
• Life Sciences
• Life Sciences
• ME Seminar
• MRC Biostatistics Unit Seminars
• Primary Care
• PublicHealth@Cambridge
• Trust & Technology Initiative - interesting events
• bld31
• my_list
• ndk22's list
• ob366-ai4er
• other talks
• primary care
• rp587

Note that ex-directory lists are not shown.