arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1902.01482cs.LGstat.ML
收藏

自举坐标搜索在多维标度中的应用

Bootstrapped Coordinate Search for Multidimensional Scaling

Department of Computer Science(计算机科学系);University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校)

Efthymios Tzinis

摘要:

在这项工作中,提出了一个统一的框架梯度自由多维标度(MDS)的基础上坐标搜索(CS)。这类算法是通用模式搜索(GPS)方法的一个实例,其避免了导数的显式计算,而是在嵌入空间的坐标步骤上搜索时评估目标函数。CSMDS框架的骨干元素是对应的概率矩阵,其对应于每个对应坐标被评估的可能性。我们提出了一个Bootstrapped的CSMDS(BS CSMDS)的实例,它提高了降低目标函数最多的方向的概率,同时也降低了所有其他坐标的相应概率。BS CSMDS设法避免不必要的功能评估,并导致比其他CSMDS替代方案显著的加速,同时还获得相同的出错率。合成和真实数据上的实验表明,BS CSMDS在各种实验设置下的性能始终优于其他CSMDS替代品。

英文摘要:

In this work, a unified framework for gradient-free Multidimensional Scaling (MDS) based on Coordinate Search (CS) is proposed. This family of algorithms is an instance of General Pattern Search (GPS) methods which avoid the explicit computation of derivatives but instead evaluate the objective function while searching on coordinate steps of the embedding space. The backbone element of CSMDS framework is the corresponding probability matrix that correspond to how likely is each corresponding coordinate to be evaluated. We propose a Bootstrapped instance of CSMDS (BS CSMDS) which enhances the probability of the direction that decreases the most the objective function while also reducing the corresponding probability of all the other coordinates. BS CSMDS manages to avoid unnecessary function evaluations and result to significant speedup over other CSMDS alternatives while also obtaining the same error rate. Experiments on both synthetic and real data reveal that BS CSMDS performs consistently better than other CSMDS alternatives under various experimental setups.

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。

↑