Convex Optimal Uncertainty Quantification

摘要: Optimal uncertainty quantification (OUQ) is a framework for numerical extreme-case analysis of stochastic systems with imperfect knowledge the underlying probability distribution. This paper presents sufficient conditions under which an OUQ problem can be reformulated as finite-dimensional convex optimization problem, efficient solutions obtained. The include that objective function piecewise concave and constraints are convex. In particular, we show functions may appear in applications where defined by optimal value parameterized linear program.