arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2007.12882stat.MLcs.LG
收藏

岭函数估计中良性过拟合现象的有限样本分析

A finite sample analysis of the benign overfitting phenomenon for ridge function estimation

University of Lyon 2(里昂第二大学);University of Oxford(牛津大学);National Physical Laboratory(英国国家物理实验室);The Alan Turing Institute(艾伦·图灵研究所);FEMTO-ST Institute(FEMTO-ST研究所)

Emmanuel Caron, Stephane Chretien

AI总结:

本文针对岭型非线性模型,在有限样本和过参数化机制下分析双重下降现象,给出估计误差的精确刻画和泛化界,补充了现有线性模型良性过拟合理论。

摘要:

近期在高规模机器学习领域的大量数值实验揭示了一种相当反直觉的相变现象,该现象是样本量与模型中参数数量之比的函数。当参数数量 $p$ 接近样本量 $n$ 时,泛化误差会增大,但令人惊讶的是,在超过阈值 $p=n$ 之后,泛化误差又开始下降。这一现象由 \cite{belkin2019reconciling} 引入理论界的关注,近来得到了深入研究,尤其是针对比深度神经网络更简单的模型,例如当参数取为最小二乘问题的最小范数解时的线性模型,首先是在 $p$ 和 $n$ 趋于无穷的渐近机制下进行研究,参见 \cite{hastie2019surprises},最近则在有限维机制下,特别是针对线性模型 \cite{bartlett2020benign}、\cite{tsigler2020benign}、\cite{lecue2022geometrical} 进行了研究。在本文中,我们提出了对 \textit{ridge} 型非线性模型的有限样本分析,其中我们针对 \textit{估计问题} 和 \textit{预测问题} 研究了双重下降现象中的 \textit{过参数化机制}。我们的结果提供了对最佳估计量与真实参数之间距离的精确分析,以及一个泛化界,该泛化界补充了 \cite{bartlett2020benign} 和 \cite{chinot2020benign} 的近期工作。我们的分析基于与连续牛顿法 \cite{neuberger2007continuous} 密切相关的工具,以及对最小 $\ell_2$-范数解预测性能的精细定量分析。

英文摘要:

Recent extensive numerical experiments in high scale machine learning have allowed to uncover a quite counterintuitive phase transition, as a function of the ratio between the sample size and the number of parameters in the model. As the number of parameters $p$ approaches the sample size $n$, the generalisation error increases, but surprisingly, it starts decreasing again past the threshold $p=n$. This phenomenon, brought to the theoretical community attention in \cite{belkin2019reconciling}, has been thoroughly investigated lately, more specifically for simpler models than deep neural networks, such as the linear model when the parameter is taken to be the minimum norm solution to the least-squares problem, firstly in the asymptotic regime when $p$ and $n$ tend to infinity, see e.g. \cite{hastie2019surprises}, and recently in the finite dimensional regime and more specifically for linear models \cite{bartlett2020benign}, \cite{tsigler2020benign}, \cite{lecue2022geometrical}. In the present paper, we propose a finite sample analysis of non-linear models of \textit{ridge} type, where we investigate the \textit{overparametrised regime} of the double descent phenomenon for both the \textit{estimation problem} and the \textit{prediction} problem. Our results provide a precise analysis of the distance of the best estimator from the true parameter as well as a generalisation bound which complements recent works of \cite{bartlett2020benign} and \cite{chinot2020benign}. Our analysis is based on tools closely related to the continuous Newton method \cite{neuberger2007continuous} and a refined quantitative analysis of the performance in prediction of the minimum $\ell_2$-norm solution.

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。

↑