2 July 2019 in Operations Research
Distributionally robust optimization (DRO), a recent methodology to handle stochastic optimization problems in the presence of data, is based on robustifications of stochastic constraints that are ... We investigate the use of distributionally robust optimization (DRO) as a tractable tool to recover the asymptotic statistical guarantees provided by the central limit theorem, for maintaining the ...
1 June 2021 in Management Science
Robust optimization (RO) is a common approach to tractably obtain safeguarding solutions for optimization problems with uncertain constraints. In this paper, we study a statistical framework to integrate data into RO based on learning a prediction set using (combinations of) geometric shapes that are compatible with established RO tools and on a simple data-splitting validation step that achieves finite-sample nonparametric statistical guarantees on feasibility. We demonstrate how our required sample size to achieve feasibility at a given confidence level is independent of the dimensions of both the decision space and the probability space governing the stochasticity, and we discuss some approaches to improve the objective performances while maintaining these dimension-free statistical feasibility guarantees. This paper was accepted by Yinyu Ye, optimization.
1 January 2019 in Operations Research
Decision making using simulation analysis faces the risk of being adversely affected by insufficiently calibrated input models. This risk is especially significant when only limited historical observations of the real-world process are available, or when unexpected risk factors are suspected and require elicitation of expert opinion. In “Robust Analysis in Stochastic Simulation: Computation and Performance Guarantees”, Ghosh and Lam propose a new methodology that solves robust optimization formulations over generic simulation models to compute bounds on the worst possible performance values. Precise performance guarantees are derived for the proposed methods, and these are further illustrated with key examples from stochastic service and queueing systems.
1 December 2017 in Operations Research
A common bottleneck in evaluating extremal performance measures is that, because of their very nature, tail data are often very limited. The conventional approach selects the best probability distribution from tail data using parametric fitting, but the validity of the parametric choice can be difficult to verify. This paper describes an alternative based on the computation of worst-case bounds under the geometric premise of tail convexity, a feature shared by all common parametric tail distributions. We characterize the optimality structure of the resulting optimization problem, and demonstrate that the worst-case convex tail behavior is in a sense either extremely light tailed or extremely heavy tailed. We develop low-dimensional nonlinear programs that distinguish between the two cases and compute the worst-case bound. We numerically illustrate how the proposed approach can give more reliable performances than conventional parametric methods.The online appendix is available at https://doi.org/10.1287/opre.2017.1643 .
6 July 2022 in ACM Transactions on Modeling and Computer Simulation
We study rare-event simulation for a class of problems where the target hitting sets of interest are defined via modern machine learning tools such as neural networks and random forests. This problem is motivated from fast emerging studies on the safety evaluation of intelligent systems, robustness quantification of learning models, and other potential applications to large-scale simulation in which machine learning tools can be used to approximate complex rare-event set boundaries. We investigate an importance sampling scheme that integrates the dominating point machinery in large deviations and sequential mixed integer programming to locate the underlying dominating points. Our approach works for a range of neural network architectures including fully connected layers, rectified linear units, normalization, pooling and convolutional layers, and random forests built from standard decision trees. We provide efficiency guarantees and numerical demonstration of our approach using a classification model in the UCI Machine Learning Repository.
1 May 2022 in Operations Research
Quantifying the impact of input estimation errors in data-driven stochastic simulation often encounters substantial computational challenges due to the entanglement of Monte Carlo and input data noises. In this paper, we propose a subsampling framework to bypass this computational bottleneck, by leveraging the form of the output variance and its estimation error in terms of data size and sampling effort. Compared with standard subsampling in the literature, our motivation is distinctly to reduce the sampling complexity of the two-layer bootstrap required in simulation uncertainty quantification. Compared with standard bootstraps, our subsampling approach provably and experimentally leads to more accurate variance and confidence interval estimations under the same amount of simulation budget.