Bayesian Active Learning of (small) Quantile Sets through Expected Estimator Modification

Given a multivariate function taking deterministic and uncertain inputs, we consider the problem of estimating a quantile set: a set of deterministic inputs for which the probability that the output belongs to a specific region remains below a given threshold. To solve this problem in the context of expensive-to-evaluate black-box functions, we propose a Bayesian active learning strategy based on Gaussian process modeling. The strategy is driven by a novel sampling criterion, which belongs to a broader principle that we refer to as Expected Estimator Modification (EEM). More specifically, the strategy relies on a novel sampling criterion combined with a sequential Monte Carlo framework that enables the construction of batch-sequential designs for the efficient estimation of small quantile sets. The performance of the strategy is illustrated on several synthetic examples and an industrial application case involving the ROTOR37 compressor model.

Bayesian sequential design of computer experiments to estimate reliable sets

We consider an unknown multivariate function representing a system-such as a complex numerical simulator-taking both deterministic and uncertain inputs. Our objective is to estimate the set of deterministic inputs leading to outputs whose probability (with respect to the distribution of the uncertain inputs) to belong to a given set is controlled by a given threshold. To solve this problem, we propose a Bayesian strategy based on the Stepwise Uncertainty Reduction (SUR) principle to sequentially choose the points at which the function should be evaluated to approximate the set of interest. We illustrate its performance and interest in several numerical experiments.