arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2103.14723stat.MLcs.LG
收藏

非线性学习模型推广误差的下界

Lower Bounds on the Generalization Error of Nonlinear Learning Models

Weizmann Institute of Science(魏茨曼科学研究所)

Inbar Seroussi, Ofer Zeitouni

摘要:

本文研究了多层神经网络模型的泛化误差的下界,在该模型中,各层的大小与训练数据中的样本数相当。我们发现,无偏估计量有不可接受的性能,这种非线性网络在这种制度。在线性回归和两层网络的情况下,我们得到了一般有偏估计的显式推广下界。在线性情况下,界是渐近紧的。在非线性的情况下,我们提供了一个比较我们的边界与随机梯度下降算法的实证研究。该分析使用来自大随机矩阵理论的元素。

英文摘要:

We study in this paper lower bounds for the generalization error of models derived from multi-layer neural networks, in the regime where the size of the layers is commensurate with the number of samples in the training data. We show that unbiased estimators have unacceptable performance for such nonlinear networks in this regime. We derive explicit generalization lower bounds for general biased estimators, in the cases of linear regression and of two-layered networks. In the linear case the bound is asymptotically tight. In the nonlinear case, we provide a comparison of our bounds with an empirical study of the stochastic gradient descent algorithm. The analysis uses elements from the theory of large random matrices.

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。

↑