BriefGPT.xyz
Oct, 2024
沿着近端随机梯度下降轨迹估计鲁棒回归的泛化性能
Estimating Generalization Performance Along the Trajectory of Proximal SGD in Robust Regression
HTML
PDF
Kai Tan, Pierre C. Bellec
TL;DR
本文研究了在高维鲁棒回归问题中,通过梯度下降(GD)、随机梯度下降(SGD)及其近端变体获得的迭代结果的泛化性能。通过引入合适条件下可证明一致的估计量,我们提供了明确的泛化误差估计,并有效地确定了最小化泛化误差的最佳停止迭代。
Abstract
This paper studies the
Generalization Performance
of iterates obtained by Gradient Descent (GD),
Stochastic Gradient Descent
(SGD) and their proximal variants in high-dimensional
→