Infinite Variational Autoencoder for Semi-Supervised Learning

M. Ehsan Abbasnejad, Anthony Dick, Anton van den Hengel

CVPR 2017

doi:10.1109/CVPR.2017.90 /cvpr/2017/abbasnejad2017cvpr-infinite/

Abstract

This paper presents an infinite variational autoencoder (VAE) whose capacity adapts to suit the input data. This is achieved using a mixture model where the mixing coefficients are modeled by a Dirichlet process, allowing us to integrate over the coefficients when performing inference. Critically, this then allows us to automatically vary the number of autoencoders in the mixture based on the data. Experiments show the flexibility of our method, particularly for semi-supervised learning, where only a small number of training samples are available.

PDF CVPR Semantic Scholar

Cite

Text

Abbasnejad et al. "Infinite Variational Autoencoder for Semi-Supervised Learning." Conference on Computer Vision and Pattern Recognition, 2017. doi:10.1109/CVPR.2017.90

Markdown

[Abbasnejad et al. "Infinite Variational Autoencoder for Semi-Supervised Learning." Conference on Computer Vision and Pattern Recognition, 2017.](https://mlanthology.org/cvpr/2017/abbasnejad2017cvpr-infinite/) doi:10.1109/CVPR.2017.90

BibTeX

@inproceedings{abbasnejad2017cvpr-infinite,
  title     = {{Infinite Variational Autoencoder for Semi-Supervised Learning}},
  author    = {Abbasnejad, M. Ehsan and Dick, Anthony and van den Hengel, Anton},
  booktitle = {Conference on Computer Vision and Pattern Recognition},
  year      = {2017},
  doi       = {10.1109/CVPR.2017.90},
  url       = {https://mlanthology.org/cvpr/2017/abbasnejad2017cvpr-infinite/}
}