论文标题
基于M-Location的风险学习
Learning with risks based on M-location
论文作者
论文摘要
在这项工作中，我们研究了根据损失分布的位置和偏差定义的新风险类别，远远超出了经典的均值差异风险功能。该课程很容易作为围绕任何平稳损失的包装器实施，它承认有限样本的平稳性保证了随机梯度方法，解释和调整很简单，并与损失位置的M-估计器紧密链接，并且对测试损失分布具有显着效果。
In this work, we study a new class of risks defined in terms of the location and deviation of the loss distribution, generalizing far beyond classical mean-variance risk functions. The class is easily implemented as a wrapper around any smooth loss, it admits finite-sample stationarity guarantees for stochastic gradient methods, it is straightforward to interpret and adjust, with close links to M-estimators of the loss location, and has a salient effect on the test loss distribution.
