Tightly Robust Optimization via Empirical Domain Reduction
Yabe, Akihiro, Maehara, Takanori
Data-driven decision-making is performed by solving a parameterized optimization problem, and the optimal decision is given by an optimal solution for unknown true parameters. We often need a solution that satisfies true constraints even though these are unknown. Robust optimization is employed to obtain such a solution, where the uncertainty of the parameter is represented by an ellipsoid, and the scale of robustness is controlled by a coefficient. In this study, we propose an algorithm to determine the scale such that the solution has a good objective value and satisfies the true constraints with a given confidence probability. Under some regularity conditions, the scale obtained by our algorithm is asymptotically $O(1/\sqrt{n})$, whereas the scale obtained by a standard approach is $O(\sqrt{d/n})$. This means that our algorithm is less affected by the dimensionality of the parameters.
Feb-29-2020
- Country:
- North America > United States (0.04)
- Europe > United Kingdom
- England > Oxfordshire > Oxford (0.04)
- Genre:
- Research Report > New Finding (0.34)
- Technology: