Large Margin Nearest Neighbor Classification using Curved Mahalanobis Distances
Machine Learning
2016-09-27 v2 Computational Geometry
Computer Vision and Pattern Recognition
Abstract
We consider the supervised classification problem of machine learning in Cayley-Klein projective geometries: We show how to learn a curved Mahalanobis metric distance corresponding to either the hyperbolic geometry or the elliptic geometry using the Large Margin Nearest Neighbor (LMNN) framework. We report on our experimental results, and further consider the case of learning a mixed curved Mahalanobis distance. Besides, we show that the Cayley-Klein Voronoi diagrams are affine, and can be built from an equivalent (clipped) power diagrams, and that Cayley-Klein balls have Mahalanobis shapes with displaced centers.
Keywords
Cite
@article{arxiv.1609.07082,
title = {Large Margin Nearest Neighbor Classification using Curved Mahalanobis Distances},
author = {Frank Nielsen and Boris Muzellec and Richard Nock},
journal= {arXiv preprint arXiv:1609.07082},
year = {2016}
}
Comments
21 pages, 8 figures, 5 tables, extend ICIP 2016 paper entitled "classification With Mixtures of Curved Mahalanobis Metrics"