arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1802.06485stat.MLcs.AIcs.LG
收藏

基于稳健梯度估计的稳健估计

Robust Estimation via Robust Gradient Estimation

Carnegie Mellon University(卡内基梅隆大学)

Adarsh Prasad, Arun Sai Suggala, Sivaraman Balakrishnan, Pradeep Ravikumar

摘要:

我们提供了一个新的计算效率的估计类风险最小化。我们表明,这些估计是强大的一般统计模型:在经典的Huber ε-污染模型和重尾设置。我们的主力是一个新的强大的梯度下降的变种,我们提供的条件下,我们的梯度下降的变种提供准确的估计在一个一般的凸风险最小化问题。我们提供了具体的后果,我们的理论,线性回归,逻辑斯谛回归和估计的典型参数在指数家庭。这些结果为这些典型的统计模型提供了一些第一个计算上易处理的和可证明的鲁棒估计。最后,我们研究了我们提出的方法在合成和真实数据集上的实证性能,发现我们的方法令人信服地优于各种基线。

英文摘要:

We provide a new computationally-efficient class of estimators for risk minimization. We show that these estimators are robust for general statistical models: in the classical Huber epsilon-contamination model and in heavy-tailed settings. Our workhorse is a novel robust variant of gradient descent, and we provide conditions under which our gradient descent variant provides accurate estimators in a general convex risk minimization problem. We provide specific consequences of our theory for linear regression, logistic regression and for estimation of the canonical parameters in an exponential family. These results provide some of the first computationally tractable and provably robust estimators for these canonical statistical models. Finally, we study the empirical performance of our proposed methods on synthetic and real datasets, and find that our methods convincingly outperform a variety of baselines.

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。

↑