SqueezeFit: Label-Aware Dimensionality Reduction by Semidefinite Programming
SqueezeFit: Label-Aware Dimensionality Reduction by Semidefinite Programming
复制标题
DOI：
10.1109/tit.2019.2962681
复制
发表时间：
2020-06-01
影响因子：
2.5
通讯作者：
Villar, Soledad
中科院分区：
文献类型：
--
作者：
McWhirter, Culver;Mixon, Dustin G.;Villar, Soledad
给定高维矢量空间中的标记点，我们寻求一个低维的子空间，以便在此子空间上投射到不同标签的点之间的某些规定距离。预期的应用包括压缩分类。本文从最接近的邻居分类中汲取灵感，引入了该问题的半决赛放松。与其前任不同，这种放松可以接受理论分析，从而使我们能够从数据中恢复种植的投影操作员。
Given labeled points in a high-dimensional vector space, we seek a low-dimensional subspace such that projecting onto this subspace maintains some prescribed distance between points of differing labels. Intended applications include compressive classification. Taking inspiration from large margin nearest neighbor classification, this paper introduces a semidefinite relaxation of this problem. Unlike its predecessors, this relaxation is amenable to theoretical analysis, allowing us to provably recover a planted projection operator from the data.