arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1702.08712stat.MLcs.LG
收藏

数学稳定性和假设复杂性

Algorithmic stability and hypothesis complexity

Pompeu Fabra University(庞培法布拉大学);ICREA(加泰罗尼亚高等研究院);Barcelona Graduate School of Economics(巴塞罗那经济研究生院);Universitat Pompeu Fabra(庞培法布拉大学)

Tongliang Liu, Gábor Lugosi, Gergely Neu, Dacheng Tao

摘要:

我们引入了学习算法的算法稳定性的概念-我们称之为\emph{argument stability} -它捕获了学习算法在从中选择假设的函数的赋范空间中输出的假设的稳定性。本文的主要结果限制了任何学习算法的推广误差的参数稳定性。的界限是基于鞅不等式的Banach空间的假设属于。我们应用一般的界限来约束的性能的一些学习算法的基础上经验风险最小化和随机梯度下降。

英文摘要:

We introduce a notion of algorithmic stability of learning algorithms---that we term \emph{argument stability}---that captures stability of the hypothesis output by the learning algorithm in the normed space of functions from which hypotheses are selected. The main result of the paper bounds the generalization error of any learning algorithm in terms of its argument stability. The bounds are based on martingale inequalities in the Banach space to which the hypotheses belong. We apply the general bounds to bound the performance of some learning algorithms based on empirical risk minimization and stochastic gradient descent.

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。

↑