Learning Sparse Metrics, One Feature at a Time

Abstract

Learning distance metrics from data is a fundamental problem in machine learning and useful way to extract data-driven features by using the matrix root of a distance matrix. Finding a proper metric amounts to optimization over the cone of positive definite (PD) matrices. This optimization is difficult since restricting optimization to remain within the PD… (More)

Topics

6 Figures and Tables

Slides referencing similar topics