A probabilistic framework for multi-view feature learning with many-to-many associations via neural networks

Akifumi Okuno, Tetsuya Hada, Hidetoshi Shimodaira
Proceedings of the 35th International Conference on Machine Learning, PMLR 80:3888-3897, 2018.

Abstract

A simple framework Probabilistic Multi-view Graph Embedding (PMvGE) is proposed for multi-view feature learning with many-to-many associations so that it generalizes various existing multi-view methods. PMvGE is a probabilistic model for predicting new associations via graph embedding of the nodes of data vectors with links of their associations. Multi-view data vectors with many-to-many associations are transformed by neural networks to feature vectors in a shared space, and the probability of new association between two data vectors is modeled by the inner product of their feature vectors. While existing multi-view feature learning techniques can treat only either of many-to-many association or non-linear transformation, PMvGE can treat both simultaneously. By combining Mercer’s theorem and the universal approximation theorem, we prove that PMvGE learns a wide class of similarity measures across views. Our likelihood-based estimator enables efficient computation of non-linear transformations of data vectors in large-scale datasets by minibatch SGD, and numerical experiments illustrate that PMvGE outperforms existing multi-view methods.

Cite this Paper


BibTeX
@InProceedings{pmlr-v80-okuno18a, title = {A probabilistic framework for multi-view feature learning with many-to-many associations via neural networks}, author = {Okuno, Akifumi and Hada, Tetsuya and Shimodaira, Hidetoshi}, booktitle = {Proceedings of the 35th International Conference on Machine Learning}, pages = {3888--3897}, year = {2018}, editor = {Dy, Jennifer and Krause, Andreas}, volume = {80}, series = {Proceedings of Machine Learning Research}, month = {10--15 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v80/okuno18a/okuno18a.pdf}, url = {http://proceedings.mlr.press/v80/okuno18a.html}, abstract = {A simple framework Probabilistic Multi-view Graph Embedding (PMvGE) is proposed for multi-view feature learning with many-to-many associations so that it generalizes various existing multi-view methods. PMvGE is a probabilistic model for predicting new associations via graph embedding of the nodes of data vectors with links of their associations. Multi-view data vectors with many-to-many associations are transformed by neural networks to feature vectors in a shared space, and the probability of new association between two data vectors is modeled by the inner product of their feature vectors. While existing multi-view feature learning techniques can treat only either of many-to-many association or non-linear transformation, PMvGE can treat both simultaneously. By combining Mercer’s theorem and the universal approximation theorem, we prove that PMvGE learns a wide class of similarity measures across views. Our likelihood-based estimator enables efficient computation of non-linear transformations of data vectors in large-scale datasets by minibatch SGD, and numerical experiments illustrate that PMvGE outperforms existing multi-view methods.} }
Endnote
%0 Conference Paper %T A probabilistic framework for multi-view feature learning with many-to-many associations via neural networks %A Akifumi Okuno %A Tetsuya Hada %A Hidetoshi Shimodaira %B Proceedings of the 35th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2018 %E Jennifer Dy %E Andreas Krause %F pmlr-v80-okuno18a %I PMLR %P 3888--3897 %U http://proceedings.mlr.press/v80/okuno18a.html %V 80 %X A simple framework Probabilistic Multi-view Graph Embedding (PMvGE) is proposed for multi-view feature learning with many-to-many associations so that it generalizes various existing multi-view methods. PMvGE is a probabilistic model for predicting new associations via graph embedding of the nodes of data vectors with links of their associations. Multi-view data vectors with many-to-many associations are transformed by neural networks to feature vectors in a shared space, and the probability of new association between two data vectors is modeled by the inner product of their feature vectors. While existing multi-view feature learning techniques can treat only either of many-to-many association or non-linear transformation, PMvGE can treat both simultaneously. By combining Mercer’s theorem and the universal approximation theorem, we prove that PMvGE learns a wide class of similarity measures across views. Our likelihood-based estimator enables efficient computation of non-linear transformations of data vectors in large-scale datasets by minibatch SGD, and numerical experiments illustrate that PMvGE outperforms existing multi-view methods.
APA
Okuno, A., Hada, T. & Shimodaira, H.. (2018). A probabilistic framework for multi-view feature learning with many-to-many associations via neural networks. Proceedings of the 35th International Conference on Machine Learning, in Proceedings of Machine Learning Research 80:3888-3897 Available from http://proceedings.mlr.press/v80/okuno18a.html.

Related Material