English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Generative adversarial networks (GANs) have proven successful in imagegeneration tasks. However, GAN training is inherently unstable. Although manyworks try to stabilize it by manually modifying GAN architecture, it requiresmuch expertise. Neural architecture search (NAS) has become an attractivesolution to search GANs automatically. The early NAS-GANs search onlygenerators to reduce search complexity but lead to a sub-optimal GAN. Somerecent works try to search both generator (G) and discriminator (D), but theysuffer from the instability of GAN training. To alleviate the instability, wepropose an efficient two-stage evolutionary algorithm-based NAS framework tosearch 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 trainingstrategy, and stage-2 searches D with the optimal G found in stage-1 and adoptsthe one-to-one training and weight-resetting strategies to enhance thestability of GAN training. Both stages use the non-dominated sorting method toproduce Pareto-front architectures under multiple objectives (e.g., model size,Inception Score (IS), and Fr\'echet Inception Distance (FID)). EAGAN is appliedto the unconditional image generation task and can efficiently finish thesearch on the CIFAR-10 dataset in 1.2 GPU days. Our searched GANs achievecompetitive results (IS=8.81$\pm$0.10, FID=9.91) on the CIFAR-10 dataset andsurpass prior NAS-GANs on the STL-10 dataset (IS=10.44$\pm$0.087, FID=22.18).Source code: https://github.com/marsggbo/EAGAN.

EAGAN: Efficient Two-stage Evolutionary Architecture Search for GANs