The Generalization Error of Random Features Regression: Precise Asymptotics and the Double Descent Curve
The Generalization Error of Random Features Regression: Precise Asymptotics and the Double Descent Curve
复制标题
DOI：
10.1002/cpa.22008
复制
发表时间：
2019-08
影响因子：
3
通讯作者：
Song Mei;A. Montanari
中科院分区：
文献类型：
--
作者：
Song Mei;A. Montanari
深度学习方法在反抗传统统计心态的政权中运作。神经网络体系结构通常包含比训练样本更多的参数，并且它们可以富含参数，即使后者被纯噪声取代，它们也可以插入观察到的标签。尽管它们的复杂性很大，但相同的体系结构在实际数据上遇到了较小的概括错误。
Deep learning methods operate in regimes that defy the traditional statistical mindset. Neural network architectures often contain more parameters than training samples, and are so rich that they can interpolate the observed labels, even if the latter are replaced by pure noise. Despite their huge complexity, the same architectures achieve small generalization error on real data.