arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1802.09303math.NAcs.NA
收藏

稀疏广义本征值问题的分解算法

A Decomposition Algorithm for the Sparse Generalized Eigenvalue Problem

Center for Quantum Computing, Peng Cheng Laboratory(鹏城实验室量子计算中心);Tencent AI Lab(腾讯AI实验室);School of Data and Computer Science, Sun Yat-sen University(中山大学数据与计算机科学学院)

Ganzhao Yuan, Li Shen, Wei-Shi Zheng

AI总结:

本文提出了一种新的分解算法来解决稀疏广义本征值问题,该问题在统计学习模型中广泛存在,但因其NP难而难以解决。通过随机或交换策略寻找工作集,并在小变量子集上进行全局组合搜索,结合二分搜索和坐标下降方法解决二次分数规划子问题,从而在准确度上显著优于现有方法。

摘要:

稀疏广义本征值问题出现在许多标准和现代统计学习模型中,包括稀疏主成分分析、稀疏Fisher判别分析和稀疏广义相关分析。然而,由于该问题NP难,因此难以求解。本文提出了一种新的分解方法来解决该问题。具体而言,我们使用随机或/和交换策略来寻找工作集,并在小变量子集上进行全局组合搜索。我们考虑了二分搜索方法和坐标下降方法来解决二次分数规划子问题。此外,我们还为所提出的方法提供了理论分析。实验结果表明,所提出的方法在准确度上显著且一致地优于现有方法。

英文摘要:

The sparse generalized eigenvalue problem arises in a number of standard and modern statistical learning models, including sparse principal component analysis, sparse Fisher discriminant analysis, and sparse canonical correlation analysis. However, this problem is difficult to solve since it is NP-hard. In this paper, we consider a new decomposition method to tackle this problem. Specifically, we use random or/and swapping strategies to find a working set and perform global combinatorial search over the small subset of variables. We consider a bisection search method and a coordinate descent method for solving the quadratic fractional programming subproblem. In addition, we provide some theoretical analysis for the proposed method. Our experiments have shown that the proposed method significantly and consistently outperforms existing solutions in term of accuracy.

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。

↑