Learning Distance for Sequences by Learning a Ground Metric

Bing Su, Ying Wu
Proceedings of the 36th International Conference on Machine Learning, PMLR 97:6015-6025, 2019.

Abstract

Learning distances that operate directly on multi-dimensional sequences is challenging because such distances are structural by nature and the vectors in sequences are not independent. Generally, distances for sequences heavily depend on the ground metric between the vectors in sequences. We propose to learn the distance for sequences through learning a ground Mahalanobis metric for the vectors in sequences. The learning samples are sequences of vectors for which how the ground metric between vectors induces the overall distance is given, and the objective is that the distance induced by the learned ground metric produces large values for sequences from different classes and small values for those from the same class. We formulate the metric as a parameter of the distance, bring closer each sequence to an associated virtual sequence w.r.t. the distance to reduce the number of constraints, and develop a general iterative solution for any ground-metric-based sequence distance. Experiments on several sequence datasets demonstrate the effectiveness and efficiency of our method.

Cite this Paper

BibTeX
@InProceedings{pmlr-v97-su19b, title = {Learning Distance for Sequences by Learning a Ground Metric}, author = {Su, Bing and Wu, Ying}, booktitle = {Proceedings of the 36th International Conference on Machine Learning}, pages = {6015--6025}, year = {2019}, editor = {Chaudhuri, Kamalika and Salakhutdinov, Ruslan}, volume = {97}, series = {Proceedings of Machine Learning Research}, month = {09--15 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v97/su19b/su19b.pdf}, url = {https://proceedings.mlr.press/v97/su19b.html}, abstract = {Learning distances that operate directly on multi-dimensional sequences is challenging because such distances are structural by nature and the vectors in sequences are not independent. Generally, distances for sequences heavily depend on the ground metric between the vectors in sequences. We propose to learn the distance for sequences through learning a ground Mahalanobis metric for the vectors in sequences. The learning samples are sequences of vectors for which how the ground metric between vectors induces the overall distance is given, and the objective is that the distance induced by the learned ground metric produces large values for sequences from different classes and small values for those from the same class. We formulate the metric as a parameter of the distance, bring closer each sequence to an associated virtual sequence w.r.t. the distance to reduce the number of constraints, and develop a general iterative solution for any ground-metric-based sequence distance. Experiments on several sequence datasets demonstrate the effectiveness and efficiency of our method.} }
Endnote
%0 Conference Paper %T Learning Distance for Sequences by Learning a Ground Metric %A Bing Su %A Ying Wu %B Proceedings of the 36th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2019 %E Kamalika Chaudhuri %E Ruslan Salakhutdinov %F pmlr-v97-su19b %I PMLR %P 6015--6025 %U https://proceedings.mlr.press/v97/su19b.html %V 97 %X Learning distances that operate directly on multi-dimensional sequences is challenging because such distances are structural by nature and the vectors in sequences are not independent. Generally, distances for sequences heavily depend on the ground metric between the vectors in sequences. We propose to learn the distance for sequences through learning a ground Mahalanobis metric for the vectors in sequences. The learning samples are sequences of vectors for which how the ground metric between vectors induces the overall distance is given, and the objective is that the distance induced by the learned ground metric produces large values for sequences from different classes and small values for those from the same class. We formulate the metric as a parameter of the distance, bring closer each sequence to an associated virtual sequence w.r.t. the distance to reduce the number of constraints, and develop a general iterative solution for any ground-metric-based sequence distance. Experiments on several sequence datasets demonstrate the effectiveness and efficiency of our method.
APA
Su, B. & Wu, Y.. (2019). Learning Distance for Sequences by Learning a Ground Metric. Proceedings of the 36th International Conference on Machine Learning, in Proceedings of Machine Learning Research 97:6015-6025 Available from https://proceedings.mlr.press/v97/su19b.html.

Related Material