A Caltech Library Service

Unsupervised Controllable Generation with Self-Training

Chrysos, Grigorios G. and Kossaifi, Jean and Yu, Zhiding and Anandkumar, Anima (2021) Unsupervised Controllable Generation with Self-Training. In: 2021 International Joint Conference on Neural Networks (IJCNN). IEEE , Piscataway, NJ, pp. 1-8. ISBN 978-1-6654-3900-8.

[img] PDF (2 May 2021) - Accepted Version
See Usage Policy.


Use this Persistent URL to link to this item:


Recent generative adversarial networks (GANs) are able to generate impressive photo-realistic images. However, controllable generation with GANs remains an open research problem. Achieving controllable generation requires semantically interpretable and disentangled factors of variation. It is challenging to achieve this goal using simple fixed distributions such as Gaussian distribution. Instead, we propose an unsupervised framework to learn a distribution of latent codes that control the generator through self-training. Self-training provides an iterative feedback in the GAN training, from the discriminator to the generator, and progressively improves the proposal of the latent codes as training proceeds. The latent codes are sampled from a latent variable model that is learned in the feature space of the discriminator. We consider a normalized independent component analysis model and learn its parameters through tensor factorization of the higher-order moments. Our framework exhibits better disentanglement compared to other variants such as the variational autoencoder, and is able to discover semantically meaningful latent codes without any supervision. We empirically demonstrate on both cars and faces datasets that each group of elements in the learned code controls a mode of variation with a semantic meaning, e.g. pose or background change. We also demonstrate with quantitative metrics that our method generates better results compared to other approaches.

Item Type:Book Section
Related URLs:
URLURL TypeDescription Paper
Chrysos, Grigorios G.0000-0002-0650-1856
Kossaifi, Jean0000-0002-4445-3429
Additional Information:© 2021 IEEE. Work done during an internship at NVIDIA Research.
Funding AgencyGrant Number
Record Number:CaltechAUTHORS:20201106-120158552
Persistent URL:
Official Citation:G. G. Chrysos, J. Kossaifi, Z. Yu and A. Anandkumar, "Unsupervised Controllable Generation with Self-Training," 2021 International Joint Conference on Neural Networks (IJCNN), 2021, pp. 1-8, doi: 10.1109/IJCNN52387.2021.9534045
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:106485
Deposited By: George Porter
Deposited On:06 Nov 2020 21:59
Last Modified:20 Dec 2021 22:28

Repository Staff Only: item control page