识别矩阵中的关键条目
Identifying Influential Entries in a Matrix
AI总结:
本文提出一种概率分布,用于识别矩阵中最关键的条目,并通过理论证明在采样少量条目后可精确重建矩阵,且无需假设矩阵的无相干性。
摘要:
对于任意一个大小为m×n、秩为ρ的矩阵A,我们提出了一种针对矩阵条目的概率分布(方程(2)中的元素杠杆分数),以揭示矩阵中最关键的条目。从理论角度看,我们证明了在采样至多s=O((m+n)ρ²ln(m+n))个条目(见方程(3)中s的精确值)并基于这些分数进行采样后,解核范数最小化问题可精确重建A。据我们所知,这些是目前在不假设矩阵A无相干性的情况下,对矩阵补全问题最强的理论保证。从实验角度看,我们显示高元素杠杆分数对应的条目揭示了数据矩阵的结构特性,这些特性对领域科学家具有研究价值。
英文摘要:
For any matrix A in R^(m x n) of rank ρ, we present a probability distribution over the entries of A (the element-wise leverage scores of equation (2)) that reveals the most influential entries in the matrix. From a theoretical perspective, we prove that sampling at most s = O ((m + n) ρ^2 ln (m + n)) entries of the matrix (see eqn. (3) for the precise value of s) with respect to these scores and solving the nuclear norm minimization problem on the sampled entries, reconstructs A exactly. To the best of our knowledge, these are the strongest theoretical guarantees on matrix completion without any incoherence assumptions on the matrix A. From an experimental perspective, we show that entries corresponding to high element-wise leverage scores reveal structural properties of the data matrix that are of interest to domain scientists.
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。