BriefGPT.xyz
Jul, 2014
稀疏矩阵分解的紧凸松弛
Tight convex relaxations for sparse matrix factorization
HTML
PDF
Emile Richard, Guillaume Obozinski, Jean-Philippe Vert
TL;DR
提出了一种基于新原子范数的凸形式，用于稀疏矩阵分解问题，其中假设因子的非零元素个数是固定和已知的，可应用于稀疏PCA，子空间聚类和低秩稀疏双线性回归等。使用主动集算法解决了该凸问题。
Abstract
Based on a new
atomic norm
, we propose a new convex formulation for
sparse matrix factorization
problems in which the number of nonzero elements of the factors is assumed fixed and known. The formulation counts <
→