An Adversarial Regularisation for Semi-Supervised Training of Structured Output Neural Networks - Archive ouverte HAL Access content directly
Conference Papers Year : 2017

An Adversarial Regularisation for Semi-Supervised Training of Structured Output Neural Networks

Mateusz Kozinski
  • Function : Author
  • PersonId : 1036104
Loïc Simon

Abstract

We propose a method for semi-supervised training of structured-output neural networks. Inspired by the framework of Generative Adversarial Networks (GAN), we train a discriminator to capture the notion of a 'quality' of network output. To this end, we leverage the qualitative difference between outputs obtained on labelled training data and unannotated data. The discriminator serves as a source of error signal for unlabelled data. Initial experiments in image segmentation demonstrate that including unlabelled data with the proposed loss function into the training procedure enables attaining the same network performance as in a fully supervised scenario, while using two times less annotations.
Fichier principal
Vignette du fichier
adversarial-regularisation-nips.pdf (213.59 Ko) Télécharger le fichier
Origin Files produced by the author(s)
Loading...

Dates and versions

hal-01713323 , version 1 (20-02-2018)

Identifiers

  • HAL Id : hal-01713323 , version 1

Cite

Mateusz Kozinski, Frédéric Jurie, Loïc Simon. An Adversarial Regularisation for Semi-Supervised Training of Structured Output Neural Networks. NIPS2017, Dec 2017, Long Beach, United States. ⟨hal-01713323⟩
280 View
147 Download

Share

Gmail Mastodon Facebook X LinkedIn More