Abstract
Generative adversarial networks (GANs) have proven successful in image generation tasks. However, GAN training is inherently unstable. Although many works try to stabilize it by manually modifying GAN architecture, it requires much expertise. Neural architecture search (NAS) has become an attractive solution to search GANs automatically. The early NAS-GANs search only generators to reduce search complexity but lead to a sub-optimal GAN. Some recent works try to search both generator (G) and discriminator (D), but they suffer from the instability of GAN training. To alleviate the instability, we propose an efficient two-stage evolutionary algorithm-based NAS framework to search GANs, namely EAGAN. We decouple the search of G and D into two stages, where stage-1 searches G with a fixed D and adopts the many-to-one training strategy, and stage-2 searches D with the optimal G found in stage-1 and adopts the one-to-one training and weight-resetting strategies to enhance the stability of GAN training. Both stages use the non-dominated sorting method to produce Pareto-front architectures under multiple objectives (e.g., model size, Inception Score (IS), and Fréchet Inception Distance (FID)). EAGAN is applied to the unconditional image generation task and can efficiently finish the search on the CIFAR-10 dataset in 1.2 GPU days. Our searched GANs achieve competitive results (IS=8.81±0.10, FID=9.91) on the CIFAR-10 dataset and surpass prior NAS-GANs on the STL-10 dataset (IS=10.44±0.087, FID=22.18).
Original language | English |
---|---|
Title of host publication | Computer Vision – ECCV 2022 |
Subtitle of host publication | 17th European Conference, Tel Aviv, Israel, October 23–27, 2022, Proceedings, Part XVI |
Editors | Shai Avidan, Gabriel Brostow, Moustapha Cissé, Giovanni Maria Farinella, Tal Hassner |
Publisher | Springer Cham |
Pages | 37–53 |
Number of pages | 17 |
Edition | 1st |
ISBN (Electronic) | 9783031197871 |
ISBN (Print) | 9783031197864 |
DOIs | |
Publication status | Published - 21 Oct 2022 |
Event | 17th European Conference on Computer Vision, ECCV 2022 - Tel Aviv, Israel Duration: 23 Oct 2022 → 27 Oct 2022 https://eccv2022.ecva.net/ https://link.springer.com/conference/eccv https://link.springer.com/book/10.1007/978-3-031-19769-7 |
Publication series
Name | Lecture Notes in Computer Science |
---|---|
Volume | 13676 |
ISSN (Print) | 0302-9743 |
ISSN (Electronic) | 1611-3349 |
Name | ECCV: European Conference on Computer Vision |
---|
Conference
Conference | 17th European Conference on Computer Vision, ECCV 2022 |
---|---|
Country/Territory | Israel |
City | Tel Aviv |
Period | 23/10/22 → 27/10/22 |
Internet address |