A data-driven approach to beating SAA out-of-sample

While solutions of Distributionally Robust Optimization (DRO) problems can sometimes have a higher out-of-sample expected reward than the Sample Average Approximation (SAA), there is no guarantee. In this paper, we introduce the class of Distributionally Optimistic Optimization (DOO) models, and show that it is always possible to ``beat" SAA out-of-sample if we consider not just worst-case (DRO) models but also best-case (DOO) ones. We also show, however, that this comes at a cost: Optimistic solutions are more sensitive to model error than either worst-case or SAA optimizers, and hence are less robust.

[1]  Xu Andy Sun,et al.  Adaptive Robust Optimization for the Security Constrained Unit Commitment Problem , 2013, IEEE Transactions on Power Systems.

[2]  Vishal Gupta,et al.  Robust SAA , 2014 .

[3]  Andrew E. B. Lim,et al.  Robust Empirical Optimization is Almost the Same As Mean-Variance Optimization , 2015, Oper. Res. Lett..

[4]  Ian R. Petersen,et al.  Minimax optimal control of stochastic uncertain systems with relative entropy constraints , 2000, IEEE Trans. Autom. Control..

[5]  Viet Anh Nguyen,et al.  Wasserstein Distributionally Robust Optimization: Theory and Applications in Machine Learning , 2019, Operations Research & Management Science in the Age of Analytics.

[6]  Arkadi Nemirovski,et al.  Robust solutions of uncertain linear programs , 1999, Oper. Res. Lett..

[7]  P. Khargonekar,et al.  State-space solutions to standard H/sub 2/ and H/sub infinity / control problems , 1989 .

[8]  Peter W. Glynn,et al.  Likelihood robust optimization for data-driven problems , 2013, Computational Management Science.

[9]  John Duchi,et al.  Statistics of Robust Optimization: A Generalized Empirical Likelihood Approach , 2016, Math. Oper. Res..

[10]  A. Philpott,et al.  Improving Sample Average Approximation Using Distributional Robustness , 2021, INFORMS Journal on Optimization.

[11]  Laurent El Ghaoui,et al.  Robust Solutions to Least-Squares Problems with Uncertain Data , 1997, SIAM J. Matrix Anal. Appl..

[12]  Andrew E. B. Lim,et al.  Worst-Case Sensitivity , 2020, ArXiv.

[13]  Andrew E. B. Lim,et al.  Robust Multiarmed Bandit Problems , 2015, Manag. Sci..

[14]  Akiko Takeda,et al.  Optimistic Robust Optimization With Applications To Machine Learning , 2017, ArXiv.

[15]  John C. Duchi,et al.  Variance-based Regularization with Convex Objectives , 2016, NIPS.

[16]  Wolfram Wiesemann,et al.  Optimistic Distributionally Robust Optimization for Nonparametric Likelihood Approximation , 2019, NeurIPS.

[17]  Robustness , 2020, The Science of Quantitative Information Flow.

[18]  A. V. D. Vaart,et al.  On the Asymptotic Information Bound , 1989 .

[19]  Henry Lam,et al.  The empirical likelihood approach to quantifying uncertainty in sample average approximation , 2017, Oper. Res. Lett..

[20]  P. Khargonekar,et al.  State-space solutions to standard H2 and H∞ control problems , 1988, 1988 American Control Conference.

[21]  Andrew E. B. Lim,et al.  Calibration of Distributionally Robust Empirical Optimization Models , 2021, Oper. Res..

[22]  Henry Lam,et al.  Recovering Best Statistical Guarantees via the Empirical Divergence-Based Distributionally Robust Optimization , 2016, Oper. Res..

[23]  Henry Lam,et al.  On the Impossibility of Statistically Improving Empirical Optimization: A Second-Order Stochastic Dominance Perspective , 2021, 2105.13419.