论文标题
鲁棒学习和伪模式统计的普遍损失的非凸延长
Nonconvex Extension of Generalized Huber Loss for Robust Learning and Pseudo-Mode Statistics
论文作者
论文摘要
我们提出了伪Huber损失公式的扩展概括。我们表明，将Log-Exp转换与Logistic函数一起使用，我们可以创建一个损失，将严格凸损耗的理想属性与强大的损失函数结合在一起。通过这种公式，我们表明可以利用线性收敛算法来找到最小化器。我们进一步讨论了准分子复合损失的创建，并提供无衍生的指数收敛速率算法。
We propose an extended generalization of the pseudo Huber loss formulation. We show that using the log-exp transform together with the logistic function, we can create a loss which combines the desirable properties of the strictly convex losses with robust loss functions. With this formulation, we show that a linear convergence algorithm can be utilized to find a minimizer. We further discuss the creation of a quasi-convex composite loss and provide a derivative-free exponential convergence rate algorithm.
