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 1704.08246cs.DScs.CCcs.LG
收藏

相对误差张量低秩近似

Relative Error Tensor Low Rank Approximation

UT-Austin(得克萨斯大学奥斯汀分校);IBM Almaden(IBM阿尔马登研究中心);Columbia University(哥伦比亚大学)

Zhao Song, David P. Woodruff, Peilin Zhong

摘要:

我们考虑相对于Frobenius范数的$tensors$的相对误差低秩近似:给定阶- $q$张量$A \in \mathbb{R}^{\prod_{i=1}^q n_i}$,输出秩- $k$张量$B$,其中$\|A-B\|_F^2 \leq (1+ε)$ OPT,其中OPT $= \inf_{\textrm{rank-}k~A'} \|A-A'\|_F^2$。尽管成功地获得了矩阵的相对误差低秩近似,但张量没有这样的结果。一个结构问题是可能没有秩- $k$张量$A_k$实现上述无穷大。另一个计算问题是,张量的有效相对误差低秩近似算法将允许计算张量的秩,这是NP难的。我们通过(1)双准则和(2)参数化复杂性解决方案绕过这些问题:(1)我们给出了一个算法,该算法输出一个秩$k' = O((k/ε)^{q-1})$张量$B$,$\|A-B\|_F^2 \leq (1+ε)$ OPT在$nnz(A) + n \cdot \textrm{poly}(k/ε)$时间在真实的RAM模型。这里$nnz(A)$是$A$中非零条目的数量。(2)我们给出了一个算法,任何$δ>0$输出一个秩$k$张量$B$,$\|A-B\|_F^2 \leq (1+ε)$ OPT和运行在$ ( nnz(A) + n \cdot \textrm{poly}(k/ε) + \exp(k^2/ε) ) \cdot n^δ$时间在单位成本RAM模型。为了输出一个秩$k$张量,甚至是一个双准则的解决方案与秩$Ck$为某一常数$C > 1$,我们显示了一个$2^{Ω(k^{1-o(1)})}$的时间下界下的指数时间假设。我们的研究结果给出了第一个相对误差低秩近似张量的大量强大的错误措施,什么都不知道,以及列行和管子集的选择。我们还获得了新的结果矩阵,如$nnz(A)$ -时间CUR分解,改善以前的$nnz(A)\log n$ -时间算法,这可能是独立的兴趣。

英文摘要:

We consider relative error low rank approximation of $tensors$ with respect to the Frobenius norm: given an order-$q$ tensor $A \in \mathbb{R}^{\prod_{i=1}^q n_i}$, output a rank-$k$ tensor $B$ for which $\|A-B\|_F^2 \leq (1+ε)$OPT, where OPT $= \inf_{\textrm{rank-}k~A'} \|A-A'\|_F^2$. Despite the success on obtaining relative error low rank approximations for matrices, no such results were known for tensors. One structural issue is that there may be no rank-$k$ tensor $A_k$ achieving the above infinum. Another, computational issue, is that an efficient relative error low rank approximation algorithm for tensors would allow one to compute the rank of a tensor, which is NP-hard. We bypass these issues via (1) bicriteria and (2) parameterized complexity solutions: (1) We give an algorithm which outputs a rank $k' = O((k/ε)^{q-1})$ tensor $B$ for which $\|A-B\|_F^2 \leq (1+ε)$OPT in $nnz(A) + n \cdot \textrm{poly}(k/ε)$ time in the real RAM model. Here $nnz(A)$ is the number of non-zero entries in $A$. (2) We give an algorithm for any $δ>0$ which outputs a rank $k$ tensor $B$ for which $\|A-B\|_F^2 \leq (1+ε)$OPT and runs in $ ( nnz(A) + n \cdot \textrm{poly}(k/ε) + \exp(k^2/ε) ) \cdot n^δ$ time in the unit cost RAM model. For outputting a rank-$k$ tensor, or even a bicriteria solution with rank-$Ck$ for a certain constant $C > 1$, we show a $2^{Ω(k^{1-o(1)})}$ time lower bound under the Exponential Time Hypothesis. Our results give the first relative error low rank approximations for tensors for a large number of robust error measures for which nothing was known, as well as column row and tube subset selection. We also obtain new results for matrices, such as $nnz(A)$-time CUR decompositions, improving previous $nnz(A)\log n$-time algorithms, which may be of independent interest.

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。

↑