Enhancing TripleGAN for Semi-Supervised Conditional Instance Synthesis and Classification
Authors:
- 1 June 2019
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 10083-10092
Abstract
Learning class-conditional data distributions is crucial for Generative Adversarial Networks (GAN) in semi-supervised learning. To improve both instance synthesis and classification in this setting, we propose an enhanced TripleGAN (EnhancedTGAN) model in this work. We follow the adversarial training scheme of the original TripleGAN, but completely re-design the training targets of the generator and classifier. Specifically, we adopt feature-semantics matching to enhance the generator in learning class-conditional distributions from both the aspects of statistics in the latent space and semantics consistency with respect to the generator and classifier. Since a limited amount of labeled data is not sufficient to determine satisfactory decision boundaries, we include two classifiers, and incorporate collaborative learning into our model to provide better guidance for generator training. The synthesized high-fidelity data can in turn be used for improving classifier training. In the experiments, the superior performance of our approach on multiple benchmark datasets demonstrates the effectiveness of the mutual reinforcement between the generator and classifiers in facilitating semi-supervised instance synthesis and classification.Keywords
This publication has 12 references indexed in Scilit:
- Recent Trends in Deep Learning Based Natural Language Processing [Review Article]IEEE Computational Intelligence Magazine, 2018
- Semi-Supervised Deep Learning Using Pseudo Labels for Hyperspectral Image ClassificationIEEE Transactions on Image Processing, 2017
- Max-Margin Deep Generative Models for (Semi-)Supervised LearningIEEE Transactions on Pattern Analysis and Machine Intelligence, 2017
- Semi-Supervised Image Classification With Self-Paced Cross-Task NetworksIEEE Transactions on Multimedia, 2017
- CVAE-GAN: Fine-Grained Image Generation through Asymmetric Training2017
- Infinite Variational Autoencoder for Semi-Supervised Learning2017
- Fully Convolutional Networks for Semantic SegmentationIEEE Transactions on Pattern Analysis and Machine Intelligence, 2016
- From neural PCA to deep unsupervised learning2015
- A data-driven approach to cleaning large face datasets2014
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998