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 1702.06533cs.LGstat.ML
收藏

随机典型相关分析

Stochastic Canonical Correlation Analysis

University of Chicago(芝加哥大学);Technion – Israel Institute of Technology(以色列理工学院);Toyota Technological Institute at Chicago(芝加哥丰田理工学院)

Chao Gao, Dan Garber, Nathan Srebro, Jialei Wang, Weiran Wang

摘要:

我们研究了典型相关分析(CCA),\ie的样本复杂性,估计总体典型相关所需的样本数和方向,直到任意小的错误。与温和的假设的数据分布,我们表明,为了实现$ε$ -次优性在一个适当定义的措施之间的对齐估计的典范方向和人口的解决方案,我们可以解决的经验目标,正是与$N(ε, Δ, γ)$样本,其中$Δ$是奇异值间隙的白化互协方差矩阵和$1/γ$是自协方差矩阵的条件数的上限。此外,我们可以通过抽取相同水平的样本和近似求解经验目标的随机优化算法来实现相同的学习精度;该算法基于移位和反转幂迭代,并且仅需要处理数据集的$\mathcal{O}\left(\log \frac{1}ε \right)$遍。最后,我们表明,给定典型相关的估计,移位和反转功率迭代的流版本通过仅处理一次数据,以相同的样本复杂度实现相同的学习精度。

英文摘要:

We study the sample complexity of canonical correlation analysis (CCA), \ie, the number of samples needed to estimate the population canonical correlation and directions up to arbitrarily small error. With mild assumptions on the data distribution, we show that in order to achieve $ε$-suboptimality in a properly defined measure of alignment between the estimated canonical directions and the population solution, we can solve the empirical objective exactly with $N(ε, Δ, γ)$ samples, where $Δ$ is the singular value gap of the whitened cross-covariance matrix and $1/γ$ is an upper bound of the condition number of auto-covariance matrices. Moreover, we can achieve the same learning accuracy by drawing the same level of samples and solving the empirical objective approximately with a stochastic optimization algorithm; this algorithm is based on the shift-and-invert power iterations and only needs to process the dataset for $\mathcal{O}\left(\log \frac{1}ε \right)$ passes. Finally, we show that, given an estimate of the canonical correlation, the streaming version of the shift-and-invert power iterations achieves the same learning accuracy with the same level of sample complexity, by processing the data only once.

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。

↑