arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2203.09607cs.LGstat.ML
收藏

基于复合优化的大规模分布鲁棒模型学习

Learning Distributionally Robust Models at Scale via Composite Optimization

Yale University(耶鲁大学);Wyze Labs Inc.(Wyze实验室公司);The Pennsylvania State University(宾夕法尼亚州立大学)

Farzin Haddadpour, Mohammad Mahdi Kamani, Mehrdad Mahdavi, Amin Karbasi

摘要:

为了训练对数据中的分布偏移具有鲁棒性的机器学习模型,分布鲁棒优化(DRO)已被证明是非常有效的。然而,现有的学习分布式鲁棒模型的方法要么需要解决复杂的优化问题,如半定规划或一阶方法,其收敛性与数据样本的数量呈线性关系,这阻碍了它们对大型数据集的可扩展性。在本文中,我们将展示如何不同的变种DRO只是一个有限和复合优化,我们提供可扩展的方法的实例。我们还提供了实证结果,证明了我们提出的算法相对于现有技术的有效性,以便从非常大的数据集学习鲁棒的模型。

英文摘要:

To train machine learning models that are robust to distribution shifts in the data, distributionally robust optimization (DRO) has been proven very effective. However, the existing approaches to learning a distributionally robust model either require solving complex optimization problems such as semidefinite programming or a first-order method whose convergence scales linearly with the number of data samples -- which hinders their scalability to large datasets. In this paper, we show how different variants of DRO are simply instances of a finite-sum composite optimization for which we provide scalable methods. We also provide empirical results that demonstrate the effectiveness of our proposed algorithm with respect to the prior art in order to learn robust models from very large datasets.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑