This is a PyTorch implementation of Self-Supervised GANs with Label Augmentation.
The folder BigGAN-PyTorch includes the code of the main experiments (on CIFAR-10, STL-10, and Tiny-ImageNet) in the paper, and the folder MoG includes the jupyter of the one-dimensional synthetic experiment.
Please cite our paper if you find the code useful for your research.
@inproceedings{NEURIPS2021_6cb5da35,
author = {Hou, Liang and Shen, Huawei and Cao, Qi and Cheng, Xueqi},
booktitle = {Advances in Neural Information Processing Systems},
editor = {M. Ranzato and A. Beygelzimer and Y. Dauphin and P.S. Liang and J. Wortman Vaughan},
pages = {13019--13031},
publisher = {Curran Associates, Inc.},
title = {Self-Supervised GANs with Label Augmentation},
url = {https://proceedings.neurips.cc/paper/2021/file/6cb5da3513bd26085ee3fad631ebb37a-Paper.pdf},
volume = {34},
year = {2021}
}