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 1502.05023stat.MLcs.DScs.ITcs.LGmath.IT
收藏

一种新的张量采样技术

A New Sampling Technique for Tensors

The University of Texas at Austin(德克萨斯大学奥斯汀分校)

Srinadh Bhojanapalli, Sujay Sanghavi

摘要:

在本文中,我们提出了新的技术来采样任意三阶张量,加快张量算法,最近在机器学习中获得了普及的目标。我们的主要贡献是一种新的方法,以有偏随机的方式选择可能的$n^3$元素中的仅$O(n^{1.5}/ε^2)$,同时仍然实现三个目标中的每一个:\\ { \em(a)张量稀疏化}:对于必须从任意样本形成的张量,计算很少的元素以获得良好的谱近似,以及对于任意正交张量{ \em(b)张量完成:}通过交替最小二乘法或{ \em(c)张量因式分解:}从少量样本中恢复精确的低秩张量,逼近被噪声破坏的低秩张量的因子。我们的采样可以与现有的基于张量的算法一起使用,以加快它们的速度,消除这些方法中的计算瓶颈。

英文摘要:

In this paper we propose new techniques to sample arbitrary third-order tensors, with an objective of speeding up tensor algorithms that have recently gained popularity in machine learning. Our main contribution is a new way to select, in a biased random way, only $O(n^{1.5}/ε^2)$ of the possible $n^3$ elements while still achieving each of the three goals: \\ {\em (a) tensor sparsification}: for a tensor that has to be formed from arbitrary samples, compute very few elements to get a good spectral approximation, and for arbitrary orthogonal tensors {\em (b) tensor completion:} recover an exactly low-rank tensor from a small number of samples via alternating least squares, or {\em (c) tensor factorization:} approximating factors of a low-rank tensor corrupted by noise. \\ Our sampling can be used along with existing tensor-based algorithms to speed them up, removing the computational bottleneck in these methods.

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。

↑