Surrogate modeling of stochastic simulators

Abstract

This thesis is a contribution to the surrogate modeling and the sensitivity analysis on stochastic simulators. Stochastic simulators are a particular type of computational models, they inherently contain some sources of randomness and are generally computationally prohibitive. To overcome this limitation, this manuscript proposes a method to build a surrogate model for stochastic simulators based on Karhunen-Loève expansion.

This thesis also aims to perform sensitivity analysis on such computational models. This analysis consists on quantifying the influence of the input variables onto the output of the model. In this thesis, the stochastic simulator is represented by a stochastic process, and the sensitivity analysis is then performed on the differential entropy of this process.

The proposed methods are applied to a stochastic simulator assessing the population’s exposure to radio frequency waves in a city. Randomness is an intrinsic characteristic of the stochastic city generator. Meaning that, for a set of city parameters (e.g. street width, building height and anisotropy) does not define a unique city. The context of the electromagnetic dosimetry case study is presented, and a surrogate model is built. The sensitivity analysis is then performed using the proposed method.

Keywords

Stochastic simulators, uncertainty quantification, surrogate modelling, sensitivity analysis, Karhunen-Loève expansion, electromagnetic dosimetry.

BibTeX cite

@PHDTHESIS{AzziThesis,
author = {Azzi, S.},
title = {Surrogate modeling of stochastic simulators},
school = {Institut Polytechnique de Paris, France},
year = {2020}
}