发表机构
Université Grenoble Alpes; Inria; CNRS; LIG; LJK(格勒诺布尔阿尔卑斯大学; 法国国家数字科学研究所; 法国国家科学研究中心; 格勒诺布尔信息学实验室; 洛朗·施瓦茨数学实验室)
机构由 AI 辅助整理,请以论文原文为准。AI 中文总结
针对输入数据受误差影响的约束优化问题,提出基于熵正则化和随机弗兰克 - 沃尔夫算法结合的可处理随机方法,用于解决瓦瑟斯坦分布鲁棒优化问题,并在两个经典问题上验证其优势,提供通用实用的处理方法。
AI 中文摘要
我们考虑输入数据受估计误差影响的约束优化问题。在此类情况下,瓦瑟斯坦分布鲁棒优化通过针对瓦瑟斯坦模糊集内的最坏情况分布进行优化,提供了一个减轻模型风险的原则性框架。然而,由此产生问题的数值求解仍然具有挑战性,特别是在约束和组合设置中。本文提出了一种基于两个关键要素的可处理随机方法:(i)对分布鲁棒值函数进行熵正则化,这使得计算随机梯度估计器成为可能;(ii)将这些估计器与随机弗兰克 - 沃尔夫算法相结合,使我们能够在自然处理约束的同时优化正则化的鲁棒目标。我们在两个经典优化问题,即最小二次生成树和交通分配问题上说明了该方法及其相对于经验风险最小化的优势。我们的方法提供了一种在存在约束的情况下处理瓦瑟斯坦分布鲁棒公式的通用且实用的方法。
英文摘要
We consider constrained problems in which input data are affected by errors. In such settings, Wasserstein distributionally robust optimization provides a principled framework to mitigate model risk by optimizing against worst-case data distributions within Wasserstein ambiguity sets. However, the numerical resolution of the resulting problems remains challenging, especially in constrained settings. In this paper, we provide a general, practical way to solve Wasserstein distributionally robust formulations in the presence of constraints. Our approach only requires a linear minimization oracle for the feasible set, and combines two key ingredients: (i) an entropic regularization of the distributionally robust value function, which makes it possible to compute stochastic gradient estimators, and (ii) a stochastic Frank-Wolfe algorithm, which minimizes the regularized robust objective while naturally handling constraints. We illustrate the method, its tractability, and its interests against empirical risk minimization, on two problems: the traffic assignment and the minimum quadratic spanning tree.