Back to Search
Start Over
Triple Generative Adversarial Networks.
- Source :
-
IEEE Transactions on Pattern Analysis & Machine Intelligence . Dec2022, Vol. 44 Issue Part2, p9629-9640. 12p. - Publication Year :
- 2022
-
Abstract
- We propose a unified game-theoretical framework to perform classification and conditional image generation given limited supervision. It is formulated as a three-player minimax game consisting of a generator, a classifier and a discriminator, and therefore is referred to as Triple Generative Adversarial Network (Triple-GAN). The generator and the classifier characterize the conditional distributions between images and labels to perform conditional generation and classification, respectively. The discriminator solely focuses on identifying fake image-label pairs. Theoretically, the three-player formulation guarantees consistency. Namely, under a nonparametric assumption, the unique equilibrium of the game is that the distributions characterized by the generator and the classifier converge to the data distribution. As a byproduct of the three-player formulation, Triple-GAN is flexible to incorporate different semi-supervised classifiers and GAN architectures. We evaluate Triple-GAN in two challenging settings, namely, semi-supervised learning and the extremely low data regime. In both settings, Triple-GAN can achieve excellent classification results and generate meaningful samples in a specific class simultaneously. In particular, using a commonly adopted 13-layer CNN classifier, Triple-GAN outperforms extensive semi-supervised learning methods substantially on several benchmarks no matter data augmentation is applied or not. [ABSTRACT FROM AUTHOR]
Details
- Language :
- English
- ISSN :
- 01628828
- Volume :
- 44
- Issue :
- Part2
- Database :
- Academic Search Index
- Journal :
- IEEE Transactions on Pattern Analysis & Machine Intelligence
- Publication Type :
- Academic Journal
- Accession number :
- 160711794
- Full Text :
- https://doi.org/10.1109/TPAMI.2021.3127558