Autoencoding beyond pixels using a learned similarity metric

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

  • Anders Boesen Lindbo Larsen
  • Søren Kaae Sønderby
  • Hugo Larochelle
  • Winther, Ole

We present an autoencoder that leverages learned representations to better measure similarities in data space. By combining a variational autoencoder (VAE) with a generative adversarial network (GAN) we can use learned feature representations in the GAN discriminator as basis for the VAE reconstruction objective. Thereby, we replace element-wise errors with feature-wise errors to better capture the data distribution while offering invariance towards e.g. translation. We apply our method to images of faces and show that it outperforms VAEs with element-wise similarity measures in terms of visual fidelity. Moreover, we show that the method learns an embedding in which high-level abstract visual features (e.g. wearing glasses) can be modified using simple arithmetic.

Original languageEnglish
Title of host publicationProceedings of The 33rd International Conference on Machine Learning
EditorsMaria Florina Balcan, Kilian Q. Weinberger
Number of pages9
Publication date2016
ISBN (Electronic)978-151082900-8
Publication statusPublished - 2016
Event33rd International Conference on Machine Learning - New York, United States
Duration: 19 Jun 201624 Jun 2016
Conference number: 33


Conference33rd International Conference on Machine Learning
LandUnited States
ByNew York
SeriesJMLR: Workshop and Conference Proceedings

ID: 171660204