Aras Selvi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Aras Selvi has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Wolfram Wiesemann
Imperial College Business School, Imperial College London, London SW7 2AZ, United Kingdom
3 shared publications - Dick Den Hertog
Amsterdam Business School, University of Amsterdam, 1018 TV Amsterdam, Netherlands
2 shared publications Aharon Ben-Tal
Faculty of Industrial Engineering and Management, Technion - Israel Institute of Technology, Haifa 3200003, Israel
1 shared publication- Huikang Liu
Shanghai Jiao Tong University, Shanghai 200240, China
1 shared publication Juan C. Vera
Department of Econometrics and Operations Research Tilburg University 5037 AB Tilburg The Netherlands
1 shared publication- Juan Carlos Vera
Department of Econometrics and Operations Research, Tilburg University, Tilburg, The Netherlands
1 shared publication
Latest Publications
- 9 March 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Manufacturing & Service Operations Management
It’s All in the Mix: Wasserstein Classification and Regression with Mixed Features
Problem definition: A key challenge in supervised learning is data scarcity, which can cause prediction models to overfit to the training data and perform poorly out of sample. A contemporary approach to combat overfitting is offered by distributionally robust problem formulations that consider all data-generating distributions close to the empirical distribution derived from historical samples, where “closeness” is determined by the Wasserstein distance. Although such formulations show significant promise in prediction tasks where all input features are continuous, they scale exponentially when discrete features are present. Methodology/results: We demonstrate that distributionally robust mixed-feature classification and regression problems can indeed be solved in polynomial time. Our proof relies on classical ellipsoid method-based solution schemes that do not scale well in practice. To overcome this limitation, we develop a practically efficient (yet, in the worst case, exponential-time) cutting-plane-based algorithm that admits a polynomial-time separation oracle, despite the presence of exponentially many constraints. We compare our method against alternative techniques both theoretically and empirically on standard benchmark instances. Managerial implications: Data-driven operations management problems often involve prediction models with discrete features. We develop and analyze distributionally robust prediction models that faithfully account for the presence of discrete features, and we demonstrate that our models can significantly outperform existing methods that are agnostic to the presence of discrete features both theoretically and on standard benchmark instances. Funding: This work was supported by the Engineering and Physical Sciences Research Council (EPSRC) [Grant EP/W003317/1]. Supplemental Material: The online appendix is available at https://doi.org/10.1287/msom.2023.0738 .
- 9 March 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Manufacturing & Service Operations Management
It’s All in the Mix: Wasserstein Classification and Regression with Mixed Features
Problem definition: A key challenge in supervised learning is data scarcity, which can cause prediction models to overfit to the training data and perform poorly out of sample. A contemporary approach to combat overfitting is offered by distributionally robust problem formulations that consider all data-generating distributions close to the empirical distribution derived from historical samples, where “closeness” is determined by the Wasserstein distance. Although such formulations show significant promise in prediction tasks where all input features are continuous, they scale exponentially when discrete features are present. Methodology/results: We demonstrate that distributionally robust mixed-feature classification and regression problems can indeed be solved in polynomial time. Our proof relies on classical ellipsoid method-based solution schemes that do not scale well in practice. To overcome this limitation, we develop a practically efficient (yet, in the worst case, exponential-time) cutting-plane-based algorithm that admits a polynomial-time separation oracle, despite the presence of exponentially many constraints. We compare our method against alternative techniques both theoretically and empirically on standard benchmark instances. Managerial implications: Data-driven operations management problems often involve prediction models with discrete features. We develop and analyze distributionally robust prediction models that faithfully account for the presence of discrete features, and we demonstrate that our models can significantly outperform existing methods that are agnostic to the presence of discrete features both theoretically and on standard benchmark instances. Funding: This work was supported by the Engineering and Physical Sciences Research Council (EPSRC) [Grant EP/W003317/1]. Supplemental Material: The online appendix is available at https://doi.org/10.1287/msom.2023.0738 .
- Top Cited Papers
- 1 January 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Operations Research
- Vol. 74 (1) , 356-376
Differential Privacy via Distributionally Robust Optimization
Privacy-Accuracy Trade-off from an Optimization Lens: Fix a Desired Level of Privacy, Then Maximize Accuracy In differential privacy, the de facto standard for safeguarding individual information in data analysis, noise is added to statistics to limit the disclosure of sensitive information. Greater privacy requires more noise, creating a trade-off as the added noise reduces the accuracy of the resulting statistics. Historically, researchers have addressed this by restricting themselves to families of noise mechanisms that are sufficient for a predefined privacy level and proving their performance under specific conditions. Selvi, Liu, and Wiesemann propose a novel approach that guarantees optimal accuracy for any specified privacy level. They formulate the design of privacy mechanisms as an optimization problem that minimizes the expected loss associated with the random noise mechanism while encoding differential privacy as constraints. Through detailed analyses and by leveraging tools from distributionally robust optimization, they develop an efficient optimization algorithm and derive implementable solutions with provable guarantees to solve the problem within seconds.
- 1 January 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Operations Research
- Vol. 74 (1) , 356-376
Top Cited PapersDifferential Privacy via Distributionally Robust Optimization
Privacy-Accuracy Trade-off from an Optimization Lens: Fix a Desired Level of Privacy, Then Maximize Accuracy In differential privacy, the de facto standard for safeguarding individual information in data analysis, noise is added to statistics to limit the disclosure of sensitive information. Greater privacy requires more noise, creating a trade-off as the added noise reduces the accuracy of the resulting statistics. Historically, researchers have addressed this by restricting themselves to families of noise mechanisms that are sufficient for a predefined privacy level and proving their performance under specific conditions. Selvi, Liu, and Wiesemann propose a novel approach that guarantees optimal accuracy for any specified privacy level. They formulate the design of privacy mechanisms as an optimization problem that minimizes the expected loss associated with the random noise mechanism while encoding differential privacy as constraints. Through detailed analyses and by leveraging tools from distributionally robust optimization, they develop an efficient optimization algorithm and derive implementable solutions with provable guarantees to solve the problem within seconds.
- Open Access
Wasserstein Logistic Regression with Mixed Features
Recent work has leveraged the popular distributionally robust optimization paradigm to combat overfitting in classical logistic regression. While the resulting classification scheme displays a promising performance in numerical experiments, it is inherently limited to numerical features. In this paper, we show that distributionally robust logistic regression with mixed (i.e., numerical and categorical) features, despite amounting to an optimization problem of exponential size, admits a polynomial-time solution scheme. We subsequently develop a practically efficient column-and-constraint approach that solves the problem as a sequence of polynomial-time solvable exponential conic programs. Our model retains many of the desirable theoretical features of previous works, but -- in contrast to the literature -- it does not admit an equivalent representation as a regularized logistic regression, that is, it represents a genuinely novel variant of logistic regression. We show that our method outperforms both the unregularized and the regularized logistic regression on categorical as well as mixed-feature benchmark instances.
Open AccessWasserstein Logistic Regression with Mixed Features
Recent work has leveraged the popular distributionally robust optimization paradigm to combat overfitting in classical logistic regression. While the resulting classification scheme displays a promising performance in numerical experiments, it is inherently limited to numerical features. In this paper, we show that distributionally robust logistic regression with mixed (i.e., numerical and categorical) features, despite amounting to an optimization problem of exponential size, admits a polynomial-time solution scheme. We subsequently develop a practically efficient column-and-constraint approach that solves the problem as a sequence of polynomial-time solvable exponential conic programs. Our model retains many of the desirable theoretical features of previous works, but -- in contrast to the literature -- it does not admit an equivalent representation as a regularized logistic regression, that is, it represents a genuinely novel variant of logistic regression. We show that our method outperforms both the unregularized and the regularized logistic regression on categorical as well as mixed-feature benchmark instances.
- 1 July 2022
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in INFORMS Journal on Computing
- Vol. 34 (4) , 2091-2105
Convex Maximization via Adjustable Robust Optimization
Maximizing a convex function over convex constraints is an NP-hard problem in general. We prove that such a problem can be reformulated as an adjustable robust optimization (ARO) problem in which each adjustable variable corresponds to a unique constraint of the original problem. We use ARO techniques to obtain approximate solutions to the convex maximization problem. In order to demonstrate the complete approximation scheme, we distinguish the cases in which we have just one nonlinear constraint and multiple linear constraints. Concerning the first case, we give three examples in which one can analytically eliminate the adjustable variable and approximately solve the resulting static robust optimization problem efficiently. More specifically, we show that the norm constrained log-sum-exp (geometric) maximization problem can be approximated by (convex) exponential cone optimization techniques. Concerning the second case of multiple linear constraints, the equivalent ARO problem can be represented as an adjustable robust linear optimization problem. Using linear decision rules then returns a safe approximation of the constraints. The resulting problem is a convex optimization problem, and solving this problem gives an upper bound on the global optimum value of the original problem. By using the optimal linear decision rule, we obtain a lower bound solution as well. We derive the approximation problems explicitly for quadratic maximization, geometric maximization, and sum-of-max-linear-terms maximization problems with multiple linear constraints. Numerical experiments show that, contrary to the state-of-the-art solvers, we can approximate large-scale problems swiftly with tight bounds. In several cases, we have equal upper and lower bounds, which concludes that we have global optimality guarantees in these cases.Summary of Contribution: Maximizing a convex function over a convex set is a hard optimization problem. We reformulate this problem as an optimization problem under uncertainty, which allows us to transfer the hardness of this problem from its nonconvexity to the uncertainty of the new problem. The equivalent uncertain optimization problem can be relaxed tightly by using adjustable robust optimization techniques. In addition to building a new bridge between convex maximization and robust optimization, this approach also gives us strong algorithms that improve the state-of-the-art optimization solvers both in solution time and quality for various convex maximization problems.
- 1 July 2022
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in INFORMS Journal on Computing
- Vol. 34 (4) , 2091-2105
Convex Maximization via Adjustable Robust Optimization
Maximizing a convex function over convex constraints is an NP-hard problem in general. We prove that such a problem can be reformulated as an adjustable robust optimization (ARO) problem in which each adjustable variable corresponds to a unique constraint of the original problem. We use ARO techniques to obtain approximate solutions to the convex maximization problem. In order to demonstrate the complete approximation scheme, we distinguish the cases in which we have just one nonlinear constraint and multiple linear constraints. Concerning the first case, we give three examples in which one can analytically eliminate the adjustable variable and approximately solve the resulting static robust optimization problem efficiently. More specifically, we show that the norm constrained log-sum-exp (geometric) maximization problem can be approximated by (convex) exponential cone optimization techniques. Concerning the second case of multiple linear constraints, the equivalent ARO problem can be represented as an adjustable robust linear optimization problem. Using linear decision rules then returns a safe approximation of the constraints. The resulting problem is a convex optimization problem, and solving this problem gives an upper bound on the global optimum value of the original problem. By using the optimal linear decision rule, we obtain a lower bound solution as well. We derive the approximation problems explicitly for quadratic maximization, geometric maximization, and sum-of-max-linear-terms maximization problems with multiple linear constraints. Numerical experiments show that, contrary to the state-of-the-art solvers, we can approximate large-scale problems swiftly with tight bounds. In several cases, we have equal upper and lower bounds, which concludes that we have global optimality guarantees in these cases.Summary of Contribution: Maximizing a convex function over a convex set is a hard optimization problem. We reformulate this problem as an optimization problem under uncertainty, which allows us to transfer the hardness of this problem from its nonconvexity to the uncertainty of the new problem. The equivalent uncertain optimization problem can be relaxed tightly by using adjustable robust optimization techniques. In addition to building a new bridge between convex maximization and robust optimization, this approach also gives us strong algorithms that improve the state-of-the-art optimization solvers both in solution time and quality for various convex maximization problems.
- Open Access
- 5 November 2021
- Published by Springer Nature in Mathematical Programming
- Vol. 197 (1) , 427-447
A reformulation-linearization technique for optimization over simplices
We study non-convex optimization problems over simplices. We show that for a large class of objective functions, the convex approximation obtained from the Reformulation-Linearization Technique (RLT) admits optimal solutions that exhibit a sparsity pattern. This characteristic of the optimal solutions allows us to conclude that (i) a linear matrix inequality constraint, which is often added to tighten the relaxation, is vacuously satisfied and can thus be omitted, and (ii) the number of decision variables in the RLT relaxation can be reduced from $${\mathcal {O}} (n^2)$$ to $${\mathcal {O}} (n)$$ . Taken together, both observations allow us to reduce computation times by up to several orders of magnitude. Our results can be specialized to indefinite quadratic optimization problems over simplices and extended to non-convex optimization problems over the Cartesian product of two simplices as well as specific classes of polyhedral and non-convex feasible regions. Our numerical experiments illustrate the promising performance of the proposed framework.- 5 November 2021
- Published by Springer Nature in Mathematical Programming
- Vol. 197 (1) , 427-447
Open AccessA reformulation-linearization technique for optimization over simplices
We study non-convex optimization problems over simplices. We show that for a large class of objective functions, the convex approximation obtained from the Reformulation-Linearization Technique (RLT) admits optimal solutions that exhibit a sparsity pattern. This characteristic of the optimal solutions allows us to conclude that (i) a linear matrix inequality constraint, which is often added to tighten the relaxation, is vacuously satisfied and can thus be omitted, and (ii) the number of decision variables in the RLT relaxation can be reduced from $${\mathcal {O}} (n^2)$$ to $${\mathcal {O}} (n)$$ . Taken together, both observations allow us to reduce computation times by up to several orders of magnitude. Our results can be specialized to indefinite quadratic optimization problems over simplices and extended to non-convex optimization problems over the Cartesian product of two simplices as well as specific classes of polyhedral and non-convex feasible regions. Our numerical experiments illustrate the promising performance of the proposed framework. - Open Access
- 3 July 2019
- Published by Taylor & Francis in The Engineering Economist
- Vol. 64 (3) , 275-288
Using column generation to solve extensions to the Markowitz model
We introduce a solution scheme for portfolio optimization problems with cardinality constraints. Typical portfolio optimization problems are extensions of the classical Markowitz mean–variance portfolio optimization model. We solve such types of problems using a method similar to column generation. In this scheme, the original problem is restricted to a subset of the assets resulting in a master convex quadratic problem. Then the dual information of the master problem is used in a subproblem to propose more assets to consider. We also consider other extensions to the Markowitz model to diversify the portfolio selection within given intervals for active weights.
- 3 July 2019
- Published by Taylor & Francis in The Engineering Economist
- Vol. 64 (3) , 275-288
Open AccessUsing column generation to solve extensions to the Markowitz model
We introduce a solution scheme for portfolio optimization problems with cardinality constraints. Typical portfolio optimization problems are extensions of the classical Markowitz mean–variance portfolio optimization model. We solve such types of problems using a method similar to column generation. In this scheme, the original problem is restricted to a subset of the assets resulting in a master convex quadratic problem. Then the dual information of the master problem is used in a subproblem to propose more assets to consider. We also consider other extensions to the Markowitz model to diversify the portfolio selection within given intervals for active weights.
- Open Access
Using Column Generation to Solve Extensions to the Markowitz Model
We introduce a solution scheme for portfolio optimization problems with cardinality constraints. Typical portfolio optimization problems are extensions of the classical Markowitz mean-variance portfolio optimization model. We solve such type of problems using a method similar to column generation. In this scheme, the original problem is restricted to a subset of the assets resulting in a master convex quadratic problem. Then the dual information of the master problem is used in a sub-problem to propose more assets to consider. We also consider other extensions to the Markowitz model to diversify the portfolio selection within the given intervals for active weights.
Open AccessUsing Column Generation to Solve Extensions to the Markowitz Model
We introduce a solution scheme for portfolio optimization problems with cardinality constraints. Typical portfolio optimization problems are extensions of the classical Markowitz mean-variance portfolio optimization model. We solve such type of problems using a method similar to column generation. In this scheme, the original problem is restricted to a subset of the assets resulting in a master convex quadratic problem. Then the dual information of the master problem is used in a sub-problem to propose more assets to consider. We also consider other extensions to the Markowitz model to diversify the portfolio selection within the given intervals for active weights.