Biconditional Generative Adversarial Networks for Multiview Learning with Missing Views

TL;DR

This paper introduces a novel biconditional GAN framework for multiview learning with missing views, jointly learning to generate missing data and classify observations without external resources.

Contribution

It proposes a tripartite GAN model that learns to generate missing views and classify data simultaneously, improving multiview learning with incomplete data.

Findings

Discriminator achieves high classification accuracy.

Generators learn high-quality missing views.

Method outperforms existing approaches on Reuters datasets.

Abstract

In this paper, we present a conditional GAN with two generators and a common discriminator for multiview learning problems where observations have two views, but one of them may be missing for some of the training samples. This is for example the case for multilingual collections where documents are not available in all languages. Some studies tackled this problem by assuming the existence of view generation functions to approximately complete the missing views; for example Machine Translation to translate documents into the missing languages. These functions generally require an external resource to be set and their quality has a direct impact on the performance of the learned multiview classifier over the completed training set. Our proposed approach addresses this problem by jointly learning the missing views and the multiview classifier using a tripartite game with two generators and a discriminator. Each of the generators is associated to one of the views and tries to fool the discriminator by generating the other missing view conditionally on the corresponding observed view. The discriminator then tries to identify if for an observation, one of its views is completed by one of the generators or if both views are completed along with its class. Our results on a subset of Reuters RCV1/RCV2 collections show that the discriminator achieves significant classification performance; and that the generators learn the missing views with high quality without the need of any consequent external resource.