arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2203.02790cs.LGcs.DScs.NAmath.NA
收藏

过完备张量分解的鲁棒谱算法

A Robust Spectral Algorithm for Overcomplete Tensor Decomposition

Cornell University(康奈尔大学);MIT(麻省理工学院);Harvard(哈佛大学)

Samuel B. Hopkins, Tselil Schramm, Jonathan Shi

摘要:

我们给出了分解过完备四阶张量的谱算法,只要它们的分量满足代数非退化条件,该条件对$(\mathbb{R}^d)^{\otimes 4}$上几乎所有(除了一个测度$0$的代数集)秩为$n \le d^2$的张量都成立.我们的算法是强大的对抗扰动有界谱范数。我们的算法的灵感来自于一个使用平方和半定规划层次(马,施,和Steurer STOC'16,arXiv:1610.01980),我们实现了类似的代数假设下的鲁棒性和过完备性保证。然而,我们的算法避免了半定规划,可以实现为一系列基本的线性代数运算。因此,我们得到了一个更快的运行时间比半定规划方法:我们的算法运行的时间$\tilde O(n^2d^3) \le \tilde O(d^7)$,这是次二次的输入大小$d^4$(其中,我们抑制了相关的输入张量的条件数的因素)。

英文摘要:

We give a spectral algorithm for decomposing overcomplete order-4 tensors, so long as their components satisfy an algebraic non-degeneracy condition that holds for nearly all (all but an algebraic set of measure $0$) tensors over $(\mathbb{R}^d)^{\otimes 4}$ with rank $n \le d^2$. Our algorithm is robust to adversarial perturbations of bounded spectral norm. Our algorithm is inspired by one which uses the sum-of-squares semidefinite programming hierarchy (Ma, Shi, and Steurer STOC'16, arXiv:1610.01980), and we achieve comparable robustness and overcompleteness guarantees under similar algebraic assumptions. However, our algorithm avoids semidefinite programming and may be implemented as a series of basic linear-algebraic operations. We consequently obtain a much faster running time than semidefinite programming methods: our algorithm runs in time $\tilde O(n^2d^3) \le \tilde O(d^7)$, which is subquadratic in the input size $d^4$ (where we have suppressed factors related to the condition number of the input tensor).

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。

↑