Learning Energy-Based Models with Adversarial Training
Abstract
We study a new approach to learning energy-based models (EBMs) based on adversarial training (AT). We show that (binary) AT learns a special kind of energy function that models the support of the data distribution, and the learning process is closely related to MCMC-based maximum likelihood learning of EBMs. We further propose improved techniques for generative modeling with AT, and demonstrate that this new approach is capable of generating diverse and realistic images. Aside from having competitive image generation performance to explicit EBMs, the studied approach is stable to train, is well-suited for image translation tasks, and exhibits strong out-of-distribution adversarial robustness. Our results demonstrate the viability of the AT approach to generative modeling, suggesting that AT is a competitive alternative approach to learning EBMs.
Cite
Text
Yin et al. "Learning Energy-Based Models with Adversarial Training." Proceedings of the European Conference on Computer Vision (ECCV), 2022. doi:10.1007/978-3-031-20065-6_13Markdown
[Yin et al. "Learning Energy-Based Models with Adversarial Training." Proceedings of the European Conference on Computer Vision (ECCV), 2022.](https://mlanthology.org/eccv/2022/yin2022eccv-learning/) doi:10.1007/978-3-031-20065-6_13BibTeX
@inproceedings{yin2022eccv-learning,
title = {{Learning Energy-Based Models with Adversarial Training}},
author = {Yin, Xuwang and Li, Shiying and Rohde, Gustavo K.},
booktitle = {Proceedings of the European Conference on Computer Vision (ECCV)},
year = {2022},
doi = {10.1007/978-3-031-20065-6_13},
url = {https://mlanthology.org/eccv/2022/yin2022eccv-learning/}
}