# A Unified Generative Adversarial Network Training via Self-Labeling and Self-Attention

arXiv: https://arxiv.org/abs/2106.09914 · HTML: https://mlsanity.com/paper/2106.09914/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 34 references that are in the map.
- First posted on arXiv 18 Jun 2021.

## Builds on (most cited first, 23 of 23)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Momentum Contrast for Unsupervised Visual Representation Learning](https://mlsanity.com/paper/1911.05722/) (2019)
- [A Style-Based Generator Architecture for Generative Adversarial Networks](https://mlsanity.com/paper/1812.04948/) (2018)
- [Large Scale GAN Training for High Fidelity Natural Image Synthesis](https://mlsanity.com/paper/1809.11096/) (2018)
- [RandAugment: Practical automated data augmentation with a reduced search space](https://mlsanity.com/paper/1909.13719/) (2019)
- [Conditional Generative Adversarial Nets](https://mlsanity.com/paper/1411.1784/) (2014)
- [FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence](https://mlsanity.com/paper/2001.07685/) (2020)
- [Spectral Normalization for Generative Adversarial Networks](https://mlsanity.com/paper/1802.05957/) (2018)
- [Generative Adversarial Text to Image Synthesis](https://mlsanity.com/paper/1605.05396/) (2016)
- [Which Training Methods for GANs do actually Converge?](https://mlsanity.com/paper/1801.04406/) (2018)
- [Conditional Image Synthesis With Auxiliary Classifier GANs](https://mlsanity.com/paper/1610.09585/) (2016)
- [Differentiable Augmentation for Data-Efficient GAN Training](https://mlsanity.com/paper/2006.10738/) (2020)
- [ReMixMatch: Semi-Supervised Learning with Distribution Alignment and Augmentation Anchoring](https://mlsanity.com/paper/1911.09785/) (2019)
- [cGANs with Projection Discriminator](https://mlsanity.com/paper/1802.05637/) (2018)
- [A U-Net Based Discriminator for Generative Adversarial Networks](https://mlsanity.com/paper/2002.12655/) (2020)
- [Self-Supervised GANs via Auxiliary Rotation Loss](https://mlsanity.com/paper/1811.11212/) (2018)
- [AutoGAN: Neural Architecture Search for Generative Adversarial Networks](https://mlsanity.com/paper/1908.03835/) (2019)
- [Improved Consistency Regularization for GANs](https://mlsanity.com/paper/2002.04724/) (2020)
- [High-Fidelity Image Generation With Fewer Labels](https://mlsanity.com/paper/1903.02271/) (2019)
- [Self-supervised GAN: Analysis and Improvement with Multi-class Minimax Game](https://mlsanity.com/paper/1911.06997/) (2019)
- [Feature Quantization Improves GAN Training](https://mlsanity.com/paper/2004.02088/) (2020)
- [cGANs with Multi-Hinge Loss](https://mlsanity.com/paper/1912.04216/) (2019)
- [Self-labeled Conditional GANs](https://mlsanity.com/paper/2012.02162/) (2020)

## Built on by (most cited first, 1 of 1)

- [Conditional GANs with Auxiliary Discriminative Classifier](https://mlsanity.com/paper/2107.10060/) (2021)

## Abstract (arXiv)

We propose a novel GAN training scheme that can handle any level of labeling in a unified manner. Our scheme introduces a form of artificial labeling that can incorporate manually defined labels, when available, and induce an alignment between them. To define the artificial labels, we exploit the assumption that neural network generators can be trained more easily to map nearby latent vectors to data with semantic similarities, than across separate categories. We use generated data samples and their corresponding artificial conditioning labels to train a classifier. The classifier is then used to self-label real data. To boost the accuracy of the self-labeling, we also use the exponential moving average of the classifier. However, because the classifier might still make mistakes, especially at the beginning of the training, we also refine the labels through self-attention, by using the labeling of real data samples only when the classifier outputs a high classification probability score. We evaluate our approach on CIFAR-10, STL-10 and SVHN, and show that both self-labeling and self-attention consistently improve the quality of generated data. More surprisingly, we find that the proposed scheme can even outperform class-conditional GANs.