arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2010.01705cs.LGcs.DSmath.STstat.MLstat.TH
收藏

带Tsybakov噪声的半空间学习的多项式时间算法

A Polynomial Time Algorithm for Learning Halfspaces with Tsybakov Noise

UW Madison(威斯康星大学麦迪逊分校);UC San-Diego(加州大学圣地亚哥分校)

Ilias Diakonikolas, Daniel M. Kane, Vasilis Kontonis, Christos Tzamos, Nikos Zarifis

摘要:

研究了齐次半空间中存在Tsybakov噪声时PAC学习问题。在Tsybakov噪声模型中,每个样本的标签都是以一个逆向控制的概率独立翻转的,对于一部分样本,该概率可以任意接近$1/2$。{ \em我们给出了这个基本学习问题的第一个多项式时间算法。}我们的算法在任何期望的精度$ε$内学习真正的半空间,并在包括对数凹分布在内的广泛的行为良好的分布家族下取得成功。在我们的工作之前,这个问题的唯一以前的算法需要准多项式运行在$1/ε$。我们的算法采用了最近开发的减少\cite{DKTZ20b}从学习到验证候选半空间的非最优性。本文提出了一种基于多项式回归的准多项式时间证书算法。{ \em当前论文的主要技术贡献是第一个多项式时间证书算法。}从一个非平凡的热启动开始,我们的算法执行一个新的“双赢”的迭代过程,在每一步中,要么找到一个有效的证书或改善当前半空间和真实的之间的角度。我们的热启动算法的各向同性对数凹分布涉及一些分析工具,可能是更广泛的兴趣。这些措施包括一个新的有效的方法,重新加权的分布,以重新定位它和一个新的表征的程度- $2$周参数的频谱。

英文摘要:

We study the problem of PAC learning homogeneous halfspaces in the presence of Tsybakov noise. In the Tsybakov noise model, the label of every sample is independently flipped with an adversarially controlled probability that can be arbitrarily close to $1/2$ for a fraction of the samples. {\em We give the first polynomial-time algorithm for this fundamental learning problem.} Our algorithm learns the true halfspace within any desired accuracy $ε$ and succeeds under a broad family of well-behaved distributions including log-concave distributions. Prior to our work, the only previous algorithm for this problem required quasi-polynomial runtime in $1/ε$. Our algorithm employs a recently developed reduction \cite{DKTZ20b} from learning to certifying the non-optimality of a candidate halfspace. This prior work developed a quasi-polynomial time certificate algorithm based on polynomial regression. {\em The main technical contribution of the current paper is the first polynomial-time certificate algorithm.} Starting from a non-trivial warm-start, our algorithm performs a novel "win-win" iterative process which, at each step, either finds a valid certificate or improves the angle between the current halfspace and the true one. Our warm-start algorithm for isotropic log-concave distributions involves a number of analytic tools that may be of broader interest. These include a new efficient method for reweighting the distribution in order to recenter it and a novel characterization of the spectrum of the degree-$2$ Chow parameters.

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。

↑