Distributionally Robust Optimization: A Review

By Hamed Rahimian, Sanjay Mehrotra. 2019

682 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Mathematical Optimization

Citations in the corpus

Adversarial Training for High-Stakes Reliability (2022, background): “Distributionally robust optimization We think our work is also related to the general project of robust optimization, especially distributionally robust optimization [43], which seeks to preserve the performance of machine learning models under distributional shift.”
Improving Task-free Continual Learning by Distributionally Robust Memory Evolution (2022, background): “Distributionally robust optimization (DRO) is an effective optimization framework to handle decision-making under uncertainty (Rahimian & Mehrotra, 2019).”
Generalizing to Unseen Domains: A Survey on Domain Generalization (2021, background): “To optimize the worst-case distribution scenario, Sagawa et al. [213] proposed a GroupDRO algo-rithm that requires explicit group annotation of the samples.”
Meta-learning without data via Wasserstein distributionally-robust model fusion (2022, background): “Distributionally Robust Optimization (DRO) [Rahimian and Mehrotra, 2019, Sinha et al., 2018] is an effective optimization framework for handling decision-making under uncertainty.”
Hybrid Variance-Reduced SGD Algorithms For Minimax Problems with Nonconvex-Linear Function (2020, background): “, in game theory, robust optimization, distributionally robust optimization, and generative adversarial nets (GANs) [4, 10, 12, 30].”
Stochastic optimization models for location and inventory prepositioning of disaster relief supplies (2020, methodology): “We refer to Rahimian and Mehrotra (2019) for a comprehensive survey of DRO.”
An Online Learning Approach to Interpolation and Extrapolation in Domain Generalization (2021, background): “…or label shift (Widmer & Kubat, 1996; Bickel et al., 2009; Lipton et al., 2018); another approach is Distributionally Robust Optimization (DRO), where the test distribution is assumed to lie in some uncertainty set around the training distribution (Bagnell, 2005; Rahimian & Mehrotra, 2019).”