arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 0902.3347stat.ML
收藏

核偏最小二乘回归加速的Lanczos逼近

Lanczos Approximations for the Speedup of Kernel Partial Least Squares Regression

Nicole Kraemer, Masashi Sugiyama, Mikio Braun

摘要:

核偏最小二乘(KPLS)计算拟合的运行时间是样本数量的二次方。然而,获得灵敏度测量作为用于模型选择的自由度或用于更详细分析的置信区间的必要性需要三次运行时间,因此构成了现实世界数据分析中的计算瓶颈。我们提出了一种新的算法KPLS不仅计算(a)的配合,但也(b)其近似自由度和(c)在二次运行时的误差条。该算法利用核PLS和Lanczos算法之间的密切联系来近似对称矩阵的特征值,并使用这种近似来计算二次运行时的核矩阵的功率的迹。

英文摘要:

The runtime for Kernel Partial Least Squares (KPLS) to compute the fit is quadratic in the number of examples. However, the necessity of obtaining sensitivity measures as degrees of freedom for model selection or confidence intervals for more detailed analysis requires cubic runtime, and thus constitutes a computational bottleneck in real-world data analysis. We propose a novel algorithm for KPLS which not only computes (a) the fit, but also (b) its approximate degrees of freedom and (c) error bars in quadratic runtime. The algorithm exploits a close connection between Kernel PLS and the Lanczos algorithm for approximating the eigenvalues of symmetric matrices, and uses this approximation to compute the trace of powers of the kernel matrix in quadratic runtime.

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。

↑