Joint Dimensionality Reduction and Metric Learning: A Geometric Take

Mehrtash Harandi, Mathieu Salzmann, Richard Hartley
Proceedings of the 34th International Conference on Machine Learning, PMLR 70:1404-1413, 2017.

Abstract

To be tractable and robust to data noise, existing metric learning algorithms commonly rely on PCA as a pre-processing step. How can we know, however, that PCA, or any other specific dimensionality reduction technique, is the method of choice for the problem at hand? The answer is simple: We cannot! To address this issue, in this paper, we develop a Riemannian framework to jointly learn a mapping performing dimensionality reduction and a metric in the induced space. Our experiments evidence that, while we directly work on high-dimensional features, our approach yields competitive runtimes with and higher accuracy than state-of-the-art metric learning algorithms.

Cite this Paper

BibTeX
@InProceedings{pmlr-v70-harandi17a, title = {Joint Dimensionality Reduction and Metric Learning: A Geometric Take}, author = {Mehrtash Harandi and Mathieu Salzmann and Richard Hartley}, booktitle = {Proceedings of the 34th International Conference on Machine Learning}, pages = {1404--1413}, year = {2017}, editor = {Precup, Doina and Teh, Yee Whye}, volume = {70}, series = {Proceedings of Machine Learning Research}, month = {06--11 Aug}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v70/harandi17a/harandi17a.pdf}, url = {https://proceedings.mlr.press/v70/harandi17a.html}, abstract = {To be tractable and robust to data noise, existing metric learning algorithms commonly rely on PCA as a pre-processing step. How can we know, however, that PCA, or any other specific dimensionality reduction technique, is the method of choice for the problem at hand? The answer is simple: We cannot! To address this issue, in this paper, we develop a Riemannian framework to jointly learn a mapping performing dimensionality reduction and a metric in the induced space. Our experiments evidence that, while we directly work on high-dimensional features, our approach yields competitive runtimes with and higher accuracy than state-of-the-art metric learning algorithms.} }
Endnote
%0 Conference Paper %T Joint Dimensionality Reduction and Metric Learning: A Geometric Take %A Mehrtash Harandi %A Mathieu Salzmann %A Richard Hartley %B Proceedings of the 34th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2017 %E Doina Precup %E Yee Whye Teh %F pmlr-v70-harandi17a %I PMLR %P 1404--1413 %U https://proceedings.mlr.press/v70/harandi17a.html %V 70 %X To be tractable and robust to data noise, existing metric learning algorithms commonly rely on PCA as a pre-processing step. How can we know, however, that PCA, or any other specific dimensionality reduction technique, is the method of choice for the problem at hand? The answer is simple: We cannot! To address this issue, in this paper, we develop a Riemannian framework to jointly learn a mapping performing dimensionality reduction and a metric in the induced space. Our experiments evidence that, while we directly work on high-dimensional features, our approach yields competitive runtimes with and higher accuracy than state-of-the-art metric learning algorithms.
APA
Harandi, M., Salzmann, M. & Hartley, R.. (2017). Joint Dimensionality Reduction and Metric Learning: A Geometric Take. Proceedings of the 34th International Conference on Machine Learning, in Proceedings of Machine Learning Research 70:1404-1413 Available from https://proceedings.mlr.press/v70/harandi17a.html.

Related Material