arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1104.1672math.PRcs.LGstat.ML
收藏

随机矩阵和的无尾不等式

Dimension-free tail inequalities for sums of random matrices

Rutgers University(罗格斯大学);University of Pennsylvania(宾夕法尼亚大学)

Daniel Hsu, Sham M. Kakade, Tong Zhang

摘要:

本文导出了与显式矩阵维数无关的随机矩阵和的指数尾不等式。这些都类似于矩阵版本的Bauffff界和伯恩斯坦不等式,除了显式矩阵维数被一个迹量所取代,即使当维数很大或无限大时,这个迹量也可以很小。给出了主成分分析和近似矩阵乘法的一些应用以说明新界的实用性。

英文摘要:

We derive exponential tail inequalities for sums of random matrices with no dependence on the explicit matrix dimensions. These are similar to the matrix versions of the Chernoff bound and Bernstein inequality except with the explicit matrix dimensions replaced by a trace quantity that can be small even when the dimension is large or infinite. Some applications to principal component analysis and approximate matrix multiplication are given to illustrate the utility of the new bounds.

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。

↑