Data-Driven Chance Constrained Programs over Wasserstein Balls

@article{Chen2018DataDrivenCC,
  title={Data-Driven Chance Constrained Programs over Wasserstein Balls},
  author={Zhi Chen and Daniel Kuhn and Wolfram Wiesemann},
  journal={Operations Research},
  year={2018}
}
In the era of modern business analytics, data-driven optimization has emerged as a popular modeling paradigm to transform data into decisions. By constructing an ambiguity set of the potential data-generating distributions and subsequently hedging against all member distributions within this ambiguity set, data-driven optimization effectively combats the ambiguity with which real-life data sets are plagued. Chen et al. (2022) study data-driven, chance-constrained programs in which a decision… 
[PDF] Semantic Reader
88 Citations
Highly Influential Citations
10
Background Citations
55
Methods Citations
26
Results Citations
3

Figures and Tables from this paper

88 Citations

Data-Driven Optimization with Distributionally Robust Second Order Stochastic Dominance Constraints

The empirical results show that, with a proper adjustment of the size of the Wasserstein ball, DRSSDCP can reach acceptable out-of-sample feasibility yet still generating strictly better performance than what is achieved by the reference strategy.

Finite-Sample Guarantees for Wasserstein Distributionally Robust Optimization: Breaking the Curse of Dimensionality

  • Rui Gao
  • Computer Science
    Operations Research
  • 2022
This paper develops the first finite-sample guarantee without suffering from the curse of dimensionality, which describes how the out-of-sample performance of a robust solution depends on the sample size, dimension of the uncertainty, and the complexity of the loss function class in a nearly optimal way.

Chance-constrained optimization under limited distributional information: A review of reformulations based on sampling and distributional robustness

Statistical Analysis of Wasserstein Distributionally Robust Estimators

This tutorial considers statistical methods which invoke a min-max distributionally robust formulation to extract good out-of-sample performance in data-driven optimization and learning problems and presents a central limit theorem for the DRO estimator and a recipe for constructing compatible confidence regions that are useful for uncertainty quantification.

Data-driven distributionally robust chance-constrained optimization with Wasserstein metric

The proposed algorithmic and reformulation framework applies to all types of distributionally robust chance-constrained optimization problems subjected to individual as well as joint chance constraints, and under two types of uncertainties, called uncertain probabilities and continuum of realizations.

DFO: A Framework for Data-driven Decision-making with Endogenous Outliers

A typical data-driven stochastic program aims to seek the best decision that minimizes the sum of a deterministic cost function and an expected recourse function under a given distribution. Recently,…

Consistency of Distributionally Robust Risk- and Chance-Constrained Optimization Under Wasserstein Ambiguity Sets

If the samples are drawn independently from an underlying distribution and the problems satisfy suitable technical assumptions, then the optimal value and optimizers of the distributionally robust versions of these problems converge to the respective quantities of the original problems, as the sample size increases.

Parametric Scenario Optimization under Limited Data: A Distributionally Robust Optimization View

This paper studies how to leverage parametric information and the power of Monte Carlo simulation to obtain feasible solutions for small-data situations by making use of a distributionally robust optimization (DRO) formulation that translates the data size requirement into a Monte Carlo sample size requirement drawn from what is called a generating distribution.

DFO: A Robust Framework for Data-driven Decision-making with Outliers

This paper shows that DFO recovers many robust statistics and can be truly “robust” in the presence of outliers, and proposes a notion of decision outlier robustness to properly select a DFO framework for the data-driven optimization with outliers.

Enhanced Wasserstein Distributionally Robust OPF With Dependence Structure and Support Information

This paper applies a moment-metric-based distributionally robust optimization, which includes a constraint on the second-order moment of uncertainty, and enhances the proposed model by adding dependence structure (correlation) and support information.
...

References

SHOWING 1-10 OF 84 REFERENCES

Distributionally Robust Optimization Under Moment Uncertainty with Application to Data-Driven Problems

This paper proposes a model that describes uncertainty in both the distribution form (discrete, Gaussian, exponential, etc.) and moments (mean and covariance matrix) and demonstrates that for a wide range of cost functions the associated distributionally robust stochastic program can be solved efficiently.

Data-driven chance constrained stochastic program

This paper derives an equivalent reformulation for DCC and shows that it is equivalent to a classical chance constraint with a perturbed risk level, and analyzes the relationship between the conservatism of D CC and the size of historical data, which can help indicate the value of data.

Wasserstein Distributionally Robust Optimization: Theory and Applications in Machine Learning

This tutorial argues that Wasserstein distributionally robust optimization has interesting ramifications for statistical learning and motivates new approaches for fundamental learning tasks such as classification, regression, maximum likelihood estimation or minimum mean square error estimation, among others.

Data-driven distributionally robust chance-constrained optimization with Wasserstein metric

The proposed algorithmic and reformulation framework applies to all types of distributionally robust chance-constrained optimization problems subjected to individual as well as joint chance constraints, and under two types of uncertainties, called uncertain probabilities and continuum of realizations.

Data-driven risk-averse stochastic optimization with Wasserstein metric

Data-driven robust optimization

This work proposes a novel schema for utilizing data to design uncertainty sets for robust optimization using statistical hypothesis tests, and shows that data-driven sets significantly outperform traditional robust optimization techniques whenever data is available.

On distributionally robust chance constrained programs with Wasserstein distance

  • Weijun Xie
  • Computer Science
    Mathematical Programming
  • 2019
It is shown that a DRCCP can be reformulated as a conditional value-at-risk constrained optimization problem, and thus admits tight inner and outer approximations and a big-M free formulation.

Distributionally Robust Stochastic Optimization with Wasserstein Distance

This paper derives a strong duality reformulation of the corresponding DRSO problem and construct approximate worst-case distributions explicitly via the first-order optimality conditions of the dual problem, and shows that data-driven DRSo problems can be approximated to any accuracy by robust optimization problems, and thereby many D RSO problems become tractable by using tools from robust optimization.

Strong Formulations for Distributionally Robust Chance-Constrained Programs with Left-Hand Side Uncertainty Under Wasserstein Ambiguity

This work exploits the connection between nominal chance-constrained programs and DR-CCP to obtain strong formulations with significant enhancements, and proposes an exponential class of inequalities that can be separated efficiently within a branch-and-cut framework.

Recovering Best Statistical Guarantees via the Empirical Divergence-Based Distributionally Robust Optimization

  • H. Lam
  • Computer Science
    Oper. Res.
  • 2019
This work investigates the use of distributionally robust optimization as a tractable tool to recover the asymptotic statistical guarantees provided by the Central Limit Theorem, and shows that using empirically defined Burg-entropy divergence balls to construct the DRO can attain such guarantees.
...