arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2102.01570cs.LGcs.CRcs.DSstat.ML
收藏

对称稀疏布尔矩阵分解及其应用

Symmetric Sparse Boolean Matrix Factorization and Applications

UC Berkeley(加州大学伯克利分校);Adobe Research(Adobe 研究院);Columbia University(哥伦比亚大学);UT Austin(德克萨斯大学奥斯汀分校)

Sitan Chen, Zhao Song, Runzhou Tao, Ruizhe Zhang

摘要:

在这项工作中,我们研究了一个变种的非负矩阵分解,我们希望找到一个对称分解的一个给定的输入矩阵到一个稀疏的,布尔矩阵。从形式上讲,给定$\mathbf{M}\in\mathbb{Z}^{m\times m}$,我们希望找到$\mathbf{W}\in\{0,1\}^{m\times r}$,使得$\| \mathbf{M} - \mathbf{W}\mathbf{W}^\top \|_0$在所有$\mathbf{W}$中最小化,其中每行都是$k$ -稀疏的。事实证明,这个问题与许多问题密切相关,如从线图中恢复超图,以及私人神经网络训练的重建攻击。由于这个问题是很难在最坏的情况下,我们研究了一个自然的平均情况下的变种,出现在这些重建攻击的上下文中:$\mathbf{M} = \mathbf{W}\mathbf{W}^{\top}$为$\mathbf{W}$一个随机布尔矩阵与$k$ -稀疏行,目标是恢复$\mathbf{W}$列置换。等价地,这可以被认为是从其线图恢复一致随机$k$ -一致超图。我们的主要结果是这个问题的多项式时间算法的基础上引导高阶信息$\mathbf{W}$,然后分解一个适当的张量。在我们的分析,这可能是独立的利益,关键成分是要表明,这样的矩阵$\mathbf{W}$有充分的列秩与高概率尽快$m = \widetildeΩ(r)$,这是我们使用的工具,从Littlewood-Offord理论和估计二进制Krawtchouk多项式。

英文摘要:

In this work, we study a variant of nonnegative matrix factorization where we wish to find a symmetric factorization of a given input matrix into a sparse, Boolean matrix. Formally speaking, given $\mathbf{M}\in\mathbb{Z}^{m\times m}$, we want to find $\mathbf{W}\in\{0,1\}^{m\times r}$ such that $\| \mathbf{M} - \mathbf{W}\mathbf{W}^\top \|_0$ is minimized among all $\mathbf{W}$ for which each row is $k$-sparse. This question turns out to be closely related to a number of questions like recovering a hypergraph from its line graph, as well as reconstruction attacks for private neural network training. As this problem is hard in the worst-case, we study a natural average-case variant that arises in the context of these reconstruction attacks: $\mathbf{M} = \mathbf{W}\mathbf{W}^{\top}$ for $\mathbf{W}$ a random Boolean matrix with $k$-sparse rows, and the goal is to recover $\mathbf{W}$ up to column permutation. Equivalently, this can be thought of as recovering a uniformly random $k$-uniform hypergraph from its line graph. Our main result is a polynomial-time algorithm for this problem based on bootstrapping higher-order information about $\mathbf{W}$ and then decomposing an appropriate tensor. The key ingredient in our analysis, which may be of independent interest, is to show that such a matrix $\mathbf{W}$ has full column rank with high probability as soon as $m = \widetildeΩ(r)$, which we do using tools from Littlewood-Offord theory and estimates for binary Krawtchouk polynomials.

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。

↑