Classification by Re-generation: Towards Classification Based on Variational Inference

As Deep Neural Networks (DNNs) are considered the state-of-the-art in many classification tasks, the question of their semantic generalizations has been raised. To address semantic interpretability of learned features, we introduce a novel idea of classification by re-generation based on variational...

Full description

Saved in:
Bibliographic Details
Published inarXiv.org
Main Authors Rezaeifar, Shideh, Taran, Olga, Voloshynovskiy, Slava
Format Paper Journal Article
LanguageEnglish
Published Ithaca Cornell University Library, arXiv.org 10.09.2018
Subjects
Classification
Computer Science - Computer Vision and Pattern Recognition
Criteria
Divergence
Encoders-Decoders
Neural networks
Semantics
Online AccessGet full text

Cover

More Information
Summary:As Deep Neural Networks (DNNs) are considered the state-of-the-art in many classification tasks, the question of their semantic generalizations has been raised. To address semantic interpretability of learned features, we introduce a novel idea of classification by re-generation based on variational autoencoder (VAE) in which a separate encoder-decoder pair of VAE is trained for each class. Moreover, the proposed architecture overcomes the scalability issue in current DNN networks as there is no need to re-train the whole network with the addition of new classes and it can be done for each class separately. We also introduce a criterion based on Kullback-Leibler divergence to reject doubtful examples. This rejection criterion should improve the trust in the obtained results and can be further exploited to reject adversarial examples.
ISSN:2331-8422
DOI:10.48550/arxiv.1809.03259