Conic Programming Reformulations of Two-Stage Distributionally Robust Linear Programs over Wasserstein Balls

Abstract

Adaptive robust optimization problems are usually solved approximately by restricting the adaptive decisions to simple parametric decision rules. However, the corresponding approximation error can be substantial. In this paper we show that two-stage robust and distributionally robust linear programs can often be reformulated exactly as conic programs that scale polynomially with the problem dimensions.Specifically, when the ambiguity set constitutes a 2-Wasserstein ball centered at a discrete distribution, then the distributionally robust linear program is equivalent to a copositive program (if the problem has complete recourse) or can be approximated arbitrarily closely by a sequence of copositive programs (if the problem has sufficiently expensive recourse). These results directly extend to the classical robust setting and motivate strong tractable approximations of two-stage problems based on semidefinite approximations of the copositive cone. We also demonstrate that the two-stage distributionally robust optimization problem is equivalent to a tractable linear program when the ambiguity set constitutes a 1-Wasserstein ball centered at a discrete distribution and there are no support constraints. arXiv:1609.07505v2 [math.OC]

On distributionally robust chance constrained programs with Wasserstein distance

“…(ii) Hanasusanto and Kuhn (2018) also proved that under the setting of Theorem 3, DRTSP with 1-Wasserstein ambiguity set is tractable. However, our formulation and required proof technique are quite different from theirs;…”

“…Similar to Hanasusanto and Kuhn (2018), we will make the following assumption throughout this paper.…”

“…showed that for DRTSP under 1−Wasserstein ambiguity set, if the recourse function can be expressed as piecewise maximum of a finite number of functions which are bi-affine in the decision variables x and the random parametersξ, then the function Z(x) has a tractable representation Hanasusanto and Kuhn (2018). extended the tractable results into DRTSP with constraint uncertainty (i.e.,ξ q is deterministic) and 1−Wasserstein ambiguity set, where the reference distance · 1 and support Ξ T = R m 2 , and proved that for general DRTSP under Wasserstein ambiguity set, it is in general NP-hard to evaluate the function Z(x).…”

Optimization under uncertainty in the era of big data and deep learning: When machine learning meets mathematical programming

“…To account for the sequential decision-making process, researchers recently developed the adaptive DRO method by incorporating recourse decision variables [123,124]. A general twostage data-driven stochastic programming model is presented in the following form:…”

Extended trust-region problems with one or two balls: exact copositive and Lagrangian relaxations

“…Copositive optimization is a special case of convex conic optimization (namely, to minimize a linear function over a cone subject to linear constraints). By now, equivalent copositive reformulations for many important problems are known, among them (non-convex, mixed-binary, fractional) quadratic optimization problems under a mild assumption [2,3,13], and some special optimization problems under uncertainty [4,18,32,37]. In particular, it has been shown in [7] that, for quadratic optimization problems with additional nonnegative constraints, copositive relaxations (and its tractable approximations) provides a tighter bound than the usual Lagrangian relaxation.…”