Guiding InfoGAN with Semi-supervision
In this paper we propose a new semi-supervised GAN architecture (ss-InfoGAN) for image synthesis that leverages information from few labels (as little as $$0.22\%$$ , max. $$10\%$$ of the dataset) to learn semantically meaningful and controllable data representations where latent variables correspon...
Saved in:
| Published in | Machine Learning and Knowledge Discovery in Databases Vol. 10534; pp. 119 - 134 |
|---|---|
| Main Authors | , , |
| Format | Book Chapter |
| Language | English |
| Published |
Switzerland
Springer International Publishing AG
2017
Springer International Publishing |
| Series | Lecture Notes in Computer Science |
| Online Access | Get full text |
Cover
Loading…
| Summary: | In this paper we propose a new semi-supervised GAN architecture (ss-InfoGAN) for image synthesis that leverages information from few labels (as little as $$0.22\%$$ , max. $$10\%$$ of the dataset) to learn semantically meaningful and controllable data representations where latent variables correspond to label categories. The architecture builds on Information Maximizing Generative Adversarial Networks (InfoGAN) and is shown to learn both continuous and categorical codes and achieves higher quality of synthetic samples compared to fully unsupervised settings. Furthermore, we show that using small amounts of labeled data speeds-up training convergence. The architecture maintains the ability to disentangle latent variables for which no labels are available. Finally, we contribute an information-theoretic reasoning on how introducing semi-supervision increases mutual information between synthetic and real data. Code related to this chapter is available at: https://github.com/spurra/ss-infogan. |
|---|---|
| Bibliography: | Electronic supplementary materialThe online version of this chapter (https://doi.org/10.1007/978-3-319-71249-9_8) contains supplementary material, which is available to authorized users. Original Abstract: In this paper we propose a new semi-supervised GAN architecture (ss-InfoGAN) for image synthesis that leverages information from few labels (as little as \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$0.22\%$$\end{document}, max. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$10\%$$\end{document} of the dataset) to learn semantically meaningful and controllable data representations where latent variables correspond to label categories. The architecture builds on Information Maximizing Generative Adversarial Networks (InfoGAN) and is shown to learn both continuous and categorical codes and achieves higher quality of synthetic samples compared to fully unsupervised settings. Furthermore, we show that using small amounts of labeled data speeds-up training convergence. The architecture maintains the ability to disentangle latent variables for which no labels are available. Finally, we contribute an information-theoretic reasoning on how introducing semi-supervision increases mutual information between synthetic and real data. Code related to this chapter is available at: https://github.com/spurra/ss-infogan. |
| ISBN: | 3319712489 9783319712482 |
| ISSN: | 0302-9743 1611-3349 |
| DOI: | 10.1007/978-3-319-71249-9_8 |