Riemannian Similarity Learning

Li Cheng
Proceedings of the 30th International Conference on Machine Learning, PMLR 28(3):540-548, 2013.

Abstract

We consider a similarity-score based paradigm to address scenarios where either the class labels are only partially revealed during learning, or the training and testing data are drawn from heterogeneous sources. The learning problem is subsequently formulated as optimization over a bilinear form of fixed rank. Our paradigm bears similarity to metric learning, where the major difference lies in its aim of learning a rectangular similarity matrix, instead of a proper metric. We tackle this problem in a Riemannian optimization framework. In particular, we consider its applications in pairwise-based action recognition, and cross-domain image-based object recognition. In both applications, the proposed algorithm produces competitive performance on respective benchmark datasets.

Cite this Paper

BibTeX
@InProceedings{pmlr-v28-cheng13, title = {Riemannian Similarity Learning}, author = {Cheng, Li}, booktitle = {Proceedings of the 30th International Conference on Machine Learning}, pages = {540--548}, year = {2013}, editor = {Dasgupta, Sanjoy and McAllester, David}, volume = {28}, number = {3}, series = {Proceedings of Machine Learning Research}, address = {Atlanta, Georgia, USA}, month = {17--19 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v28/cheng13.pdf}, url = {https://proceedings.mlr.press/v28/cheng13.html}, abstract = {We consider a similarity-score based paradigm to address scenarios where either the class labels are only partially revealed during learning, or the training and testing data are drawn from heterogeneous sources. The learning problem is subsequently formulated as optimization over a bilinear form of fixed rank. Our paradigm bears similarity to metric learning, where the major difference lies in its aim of learning a rectangular similarity matrix, instead of a proper metric. We tackle this problem in a Riemannian optimization framework. In particular, we consider its applications in pairwise-based action recognition, and cross-domain image-based object recognition. In both applications, the proposed algorithm produces competitive performance on respective benchmark datasets.} }
Endnote
%0 Conference Paper %T Riemannian Similarity Learning %A Li Cheng %B Proceedings of the 30th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2013 %E Sanjoy Dasgupta %E David McAllester %F pmlr-v28-cheng13 %I PMLR %P 540--548 %U https://proceedings.mlr.press/v28/cheng13.html %V 28 %N 3 %X We consider a similarity-score based paradigm to address scenarios where either the class labels are only partially revealed during learning, or the training and testing data are drawn from heterogeneous sources. The learning problem is subsequently formulated as optimization over a bilinear form of fixed rank. Our paradigm bears similarity to metric learning, where the major difference lies in its aim of learning a rectangular similarity matrix, instead of a proper metric. We tackle this problem in a Riemannian optimization framework. In particular, we consider its applications in pairwise-based action recognition, and cross-domain image-based object recognition. In both applications, the proposed algorithm produces competitive performance on respective benchmark datasets.
RIS
TY - CPAPER TI - Riemannian Similarity Learning AU - Li Cheng BT - Proceedings of the 30th International Conference on Machine Learning DA - 2013/05/26 ED - Sanjoy Dasgupta ED - David McAllester ID - pmlr-v28-cheng13 PB - PMLR DP - Proceedings of Machine Learning Research VL - 28 IS - 3 SP - 540 EP - 548 L1 - http://proceedings.mlr.press/v28/cheng13.pdf UR - https://proceedings.mlr.press/v28/cheng13.html AB - We consider a similarity-score based paradigm to address scenarios where either the class labels are only partially revealed during learning, or the training and testing data are drawn from heterogeneous sources. The learning problem is subsequently formulated as optimization over a bilinear form of fixed rank. Our paradigm bears similarity to metric learning, where the major difference lies in its aim of learning a rectangular similarity matrix, instead of a proper metric. We tackle this problem in a Riemannian optimization framework. In particular, we consider its applications in pairwise-based action recognition, and cross-domain image-based object recognition. In both applications, the proposed algorithm produces competitive performance on respective benchmark datasets. ER -
APA
Cheng, L.. (2013). Riemannian Similarity Learning. Proceedings of the 30th International Conference on Machine Learning, in Proceedings of Machine Learning Research 28(3):540-548 Available from https://proceedings.mlr.press/v28/cheng13.html.

Related Material