arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2303.02255cs.LGmath.OCstat.ML
收藏

学习高维单个 ReLU 神经元的有限样本分析

Finite-Sample Analysis of Learning High-Dimensional Single ReLU Neuron

Johns Hopkins University(约翰斯·霍普金斯大学);The University of Hong Kong(香港大学);University of California, Los Angeles(加利福尼亚大学洛杉矶分校);Rice University(莱斯大学);Harvard University(哈佛大学)

Jingfeng Wu, Difan Zou, Zixiang Chen, Vladimir Braverman, Quanquan Gu, Sham M. Kakade

AI总结:

本文研究了过参数化机制下高维单个 ReLU 神经元的学习问题,分析了 GLM-tron 算法并给出了风险上下界,同时证明了在此场景下 GLM-tron 的表现优于随机梯度下降(SGD)。

摘要:

本文考虑在过参数化机制下使用平方损失(即 ReLU 回归)学习单个 ReLU 神经元的问题,其中输入维度可以超过样本数量。我们分析了一种名为 GLM-tron(Kakade et al., 2011)的 Perceptron 型算法,并在正确设定和错误设定两种设置下,为高维 ReLU 回归提供了其与维度无关的风险上界。我们的风险界将现有若干结果作为特例予以恢复。此外,在正确设定设置下,我们为 GLM-tron 提供了逐实例匹配的风险下界。我们的风险上下界对可通过 GLM-tron 学习的高维 ReLU 回归问题给出了精确刻画。另一方面,对于使用对称伯努利数据的 ReLU 回归,我们提供了随机梯度下降(SGD)的一些负面结果:若模型为正确设定,在忽略常数因子的情况下,对于每个问题实例,SGD 的过量风险可证明不优于 GLM-tron;在无噪声情况下,GLM-tron 能实现较小的风险,而 SGD 在期望上不可避免地遭受常数风险。这些结果共同表明,对于高维 ReLU 回归,GLM-tron 可能优于 SGD。

英文摘要:

This paper considers the problem of learning a single ReLU neuron with squared loss (a.k.a., ReLU regression) in the overparameterized regime, where the input dimension can exceed the number of samples. We analyze a Perceptron-type algorithm called GLM-tron (Kakade et al., 2011) and provide its dimension-free risk upper bounds for high-dimensional ReLU regression in both well-specified and misspecified settings. Our risk bounds recover several existing results as special cases. Moreover, in the well-specified setting, we provide an instance-wise matching risk lower bound for GLM-tron. Our upper and lower risk bounds provide a sharp characterization of the high-dimensional ReLU regression problems that can be learned via GLM-tron. On the other hand, we provide some negative results for stochastic gradient descent (SGD) for ReLU regression with symmetric Bernoulli data: if the model is well-specified, the excess risk of SGD is provably no better than that of GLM-tron ignoring constant factors, for each problem instance; and in the noiseless case, GLM-tron can achieve a small risk while SGD unavoidably suffers from a constant risk in expectation. These results together suggest that GLM-tron might be preferable to SGD for high-dimensional ReLU regression.

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。

↑