Affinity guided Geometric Semi-Supervised Metric Learning
Abstract
In this paper, we revamp the forgotten classical Semi-Supervised Distance Metric Learning (SSDML) problem from a Riemannian geometric lens, to leverage stochastic optimization within a end-to-end deep framework. The motivation comes from the fact that apart from a few classical SSDML approaches learning a linear Mahalanobis metric, deep SSDML has not been studied. We first extend existing SSDML methods to their deep counterparts and then propose a new method to overcome their limitations. Due to the nature of constraints on our metric parameters, we leverage Riemannian optimization. Our deep SSDML method with a novel affinity propagation based triplet mining strategy outperforms its competitors.
Cite
@article{arxiv.2002.12394,
title = {Affinity guided Geometric Semi-Supervised Metric Learning},
author = {Ujjal Kr Dutta and Mehrtash Harandi and Chellu Chandra Sekhar},
journal= {arXiv preprint arXiv:2002.12394},
year = {2020}
}
Comments
Paper accepted in NeurIPS 2020 workshop on Differential Geometry meets Deep Learning