Multiview Triplet Embedding: Learning Attributes in Multiple Maps

Ehsan Amid, Antti Ukkonen
Proceedings of the 32nd International Conference on Machine Learning, PMLR 37:1472-1480, 2015.

Abstract

For humans, it is usually easier to make statements about the similarity of objects in relative, rather than absolute terms. Moreover, subjective comparisons of objects can be based on a number of different and independent attributes. For example, objects can be compared based on their shape, color, etc. In this paper, we consider the problem of uncovering these hidden attributes given a set of relative distance judgments in the form of triplets. The attribute that was used to generate a particular triplet in this set is unknown. Such data occurs, e.g., in crowdsourcing applications where the triplets are collected from a large group of workers. We propose the Multiview Triplet Embedding (MVTE) algorithm that produces a number of low-dimensional maps, each corresponding to one of the hidden attributes. The method can be used to assess how many different attributes were used to create the triplets, as well as to assess the difficulty of a distance comparison task, and find objects that have multiple interpretations in relation to the other objects.

Cite this Paper


BibTeX
@InProceedings{pmlr-v37-amid15, title = {Multiview Triplet Embedding: Learning Attributes in Multiple Maps}, author = {Amid, Ehsan and Ukkonen, Antti}, booktitle = {Proceedings of the 32nd International Conference on Machine Learning}, pages = {1472--1480}, year = {2015}, editor = {Bach, Francis and Blei, David}, volume = {37}, series = {Proceedings of Machine Learning Research}, address = {Lille, France}, month = {07--09 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v37/amid15.pdf}, url = {https://proceedings.mlr.press/v37/amid15.html}, abstract = {For humans, it is usually easier to make statements about the similarity of objects in relative, rather than absolute terms. Moreover, subjective comparisons of objects can be based on a number of different and independent attributes. For example, objects can be compared based on their shape, color, etc. In this paper, we consider the problem of uncovering these hidden attributes given a set of relative distance judgments in the form of triplets. The attribute that was used to generate a particular triplet in this set is unknown. Such data occurs, e.g., in crowdsourcing applications where the triplets are collected from a large group of workers. We propose the Multiview Triplet Embedding (MVTE) algorithm that produces a number of low-dimensional maps, each corresponding to one of the hidden attributes. The method can be used to assess how many different attributes were used to create the triplets, as well as to assess the difficulty of a distance comparison task, and find objects that have multiple interpretations in relation to the other objects.} }
Endnote
%0 Conference Paper %T Multiview Triplet Embedding: Learning Attributes in Multiple Maps %A Ehsan Amid %A Antti Ukkonen %B Proceedings of the 32nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2015 %E Francis Bach %E David Blei %F pmlr-v37-amid15 %I PMLR %P 1472--1480 %U https://proceedings.mlr.press/v37/amid15.html %V 37 %X For humans, it is usually easier to make statements about the similarity of objects in relative, rather than absolute terms. Moreover, subjective comparisons of objects can be based on a number of different and independent attributes. For example, objects can be compared based on their shape, color, etc. In this paper, we consider the problem of uncovering these hidden attributes given a set of relative distance judgments in the form of triplets. The attribute that was used to generate a particular triplet in this set is unknown. Such data occurs, e.g., in crowdsourcing applications where the triplets are collected from a large group of workers. We propose the Multiview Triplet Embedding (MVTE) algorithm that produces a number of low-dimensional maps, each corresponding to one of the hidden attributes. The method can be used to assess how many different attributes were used to create the triplets, as well as to assess the difficulty of a distance comparison task, and find objects that have multiple interpretations in relation to the other objects.
RIS
TY - CPAPER TI - Multiview Triplet Embedding: Learning Attributes in Multiple Maps AU - Ehsan Amid AU - Antti Ukkonen BT - Proceedings of the 32nd International Conference on Machine Learning DA - 2015/06/01 ED - Francis Bach ED - David Blei ID - pmlr-v37-amid15 PB - PMLR DP - Proceedings of Machine Learning Research VL - 37 SP - 1472 EP - 1480 L1 - http://proceedings.mlr.press/v37/amid15.pdf UR - https://proceedings.mlr.press/v37/amid15.html AB - For humans, it is usually easier to make statements about the similarity of objects in relative, rather than absolute terms. Moreover, subjective comparisons of objects can be based on a number of different and independent attributes. For example, objects can be compared based on their shape, color, etc. In this paper, we consider the problem of uncovering these hidden attributes given a set of relative distance judgments in the form of triplets. The attribute that was used to generate a particular triplet in this set is unknown. Such data occurs, e.g., in crowdsourcing applications where the triplets are collected from a large group of workers. We propose the Multiview Triplet Embedding (MVTE) algorithm that produces a number of low-dimensional maps, each corresponding to one of the hidden attributes. The method can be used to assess how many different attributes were used to create the triplets, as well as to assess the difficulty of a distance comparison task, and find objects that have multiple interpretations in relation to the other objects. ER -
APA
Amid, E. & Ukkonen, A.. (2015). Multiview Triplet Embedding: Learning Attributes in Multiple Maps. Proceedings of the 32nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 37:1472-1480 Available from https://proceedings.mlr.press/v37/amid15.html.

Related Material