Reduced-Rank Local Distance Metric Learning
Abstract
We propose a new method for local metric learning based on a conical combination of Mahalanobis metrics and pair-wise similarities between the data. Its formulation allows for controlling the rank of the metrics' weight matrices. We also offer a convergent algorithm for training the associated model. Experimental results on a collection of classification problems imply that the new method may offer notable performance advantages over alternative metric learning approaches that have recently appeared in the literature.
Cite
Text
Huang et al. "Reduced-Rank Local Distance Metric Learning." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2013. doi:10.1007/978-3-642-40994-3_15Markdown
[Huang et al. "Reduced-Rank Local Distance Metric Learning." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2013.](https://mlanthology.org/ecmlpkdd/2013/huang2013ecmlpkdd-reducedrank/) doi:10.1007/978-3-642-40994-3_15BibTeX
@inproceedings{huang2013ecmlpkdd-reducedrank,
title = {{Reduced-Rank Local Distance Metric Learning}},
author = {Huang, Yinjie and Li, Cong and Georgiopoulos, Michael and Anagnostopoulos, Georgios C.},
booktitle = {European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases},
year = {2013},
pages = {224-239},
doi = {10.1007/978-3-642-40994-3_15},
url = {https://mlanthology.org/ecmlpkdd/2013/huang2013ecmlpkdd-reducedrank/}
}