Low-rank geometric mean metric learning
Machine Learning
2018-06-15 v1 Machine Learning
Abstract
We propose a low-rank approach to learning a Mahalanobis metric from data. Inspired by the recent geometric mean metric learning (GMML) algorithm, we propose a low-rank variant of the algorithm. This allows to jointly learn a low-dimensional subspace where the data reside and the Mahalanobis metric that appropriately fits the data. Our results show that we compete effectively with GMML at lower ranks.
Keywords
Cite
@article{arxiv.1806.05454,
title = {Low-rank geometric mean metric learning},
author = {Mukul Bhutani and Pratik Jawanpuria and Hiroyuki Kasai and Bamdev Mishra},
journal= {arXiv preprint arXiv:1806.05454},
year = {2018}
}
Comments
Accepted to the geometry in machine learning (GiMLi) workshop at ICML 2018