arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1506.04855cs.LGstat.ML
收藏

具有高斯扰动的PCA

PCA with Gaussian perturbations

Poznań University of Technology(波兹南理工大学);University of California, Santa Cruz(加州大学圣克鲁兹分校)

Wojciech Kotłowski, Manfred K. Warmuth

摘要:

大多数机器学习都处理向量参数。理想情况下,我们希望考虑高阶信息,并利用矩阵甚至张量参数。然而,由此产生的算法通常是低效的。在这里,我们解决在线学习与矩阵参数。如果在每次试验中对当前参数矩阵进行特征分解(每次试验的代价为$O(n^3)$),通常很容易获得具有良好泛化性能的在线算法。理想情况下,我们希望避免分解,每次试验花费$O(n^2)$,即矩阵数据大小的线性时间。在运行时间和泛化性能之间存在核心权衡,这里通过在线算法的遗憾(最佳离线预测器的总增益减去在线算法的总增益)来测量。重点研究了$\mathbb{R}^n$中秩$k$主成分分析的关键矩阵问题,其中$k \ll n$ .有$O(n^3)$算法,实现最佳遗憾,但需要特征分解。我们开发了一个简单的算法,需要$O(kn^2)$每次试验的遗憾是关闭了一个小的因素$O(n^{1/4})$。该算法基于Follow the Perturbed Leader范式。它取代了完整的特征分解,在每次试验的问题,找到$k$主成分的当前协方差矩阵是由高斯噪声扰动。

英文摘要:

Most of machine learning deals with vector parameters. Ideally we would like to take higher order information into account and make use of matrix or even tensor parameters. However the resulting algorithms are usually inefficient. Here we address on-line learning with matrix parameters. It is often easy to obtain online algorithm with good generalization performance if you eigendecompose the current parameter matrix in each trial (at a cost of $O(n^3)$ per trial). Ideally we want to avoid the decompositions and spend $O(n^2)$ per trial, i.e. linear time in the size of the matrix data. There is a core trade-off between the running time and the generalization performance, here measured by the regret of the on-line algorithm (total gain of the best off-line predictor minus the total gain of the on-line algorithm). We focus on the key matrix problem of rank $k$ Principal Component Analysis in $\mathbb{R}^n$ where $k \ll n$. There are $O(n^3)$ algorithms that achieve the optimum regret but require eigendecompositions. We develop a simple algorithm that needs $O(kn^2)$ per trial whose regret is off by a small factor of $O(n^{1/4})$. The algorithm is based on the Follow the Perturbed Leader paradigm. It replaces full eigendecompositions at each trial by the problem finding $k$ principal components of the current covariance matrix that is perturbed by Gaussian noise.

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。

↑