All translations
Enter a message name below to show all available translations.
Found 3 translations.
| Name | Current message text |
|---|---|
| h English (en) | {{Term|generative adversarial network|GANs}} sparked one of the most active areas of {{Term|deep learning}} research. Within a few years of publication, thousands of {{Term|generative adversarial network|GAN}} variants were proposed, addressing training instability ({{Term|wasserstein gan|WGAN}}, {{Term|spectral normalization}}), enabling conditional generation ({{Term|conditional gan|cGAN}}, pix2pix), achieving photorealistic {{Term|column space|image}} synthesis ({{Term|stylegan}}, {{Term|biggan}}), and extending to video, 3D, and other modalities. The {{Term|adversarial training}} principle was also applied to {{Term|domain adaptation}}, {{Term|data augmentation}}, super-resolution, and text-to-{{Term|column space|image}} generation. |
| h Spanish (es) | Las {{Term|generative adversarial network|GAN}} desencadenaron una de las áreas más activas de la investigación en {{Term|deep learning|aprendizaje profundo}}. A los pocos años de la publicación, se propusieron miles de variantes de {{Term|generative adversarial network|GAN}} que abordaban la inestabilidad del entrenamiento ({{Term|wasserstein gan|WGAN}}, {{Term|spectral normalization|normalización espectral}}), permitían la generación condicional ({{Term|conditional gan|cGAN}}, pix2pix), lograban la síntesis de {{Term|column space|imágenes}} fotorrealistas ({{Term|stylegan|StyleGAN}}, {{Term|biggan|BigGAN}}) y se extendían a video, 3D y otras modalidades. El principio del {{Term|adversarial training|entrenamiento adversarial}} también se aplicó a la {{Term|domain adaptation|adaptación de dominio}}, el {{Term|data augmentation|aumento de datos}}, la superresolución y la generación de texto a {{Term|column space|imagen}}. |
| h Chinese (zh) | {{Term|generative adversarial network|GAN}} 引发了{{Term|deep learning|深度学习}}研究中最为活跃的领域之一。在论文发表后的几年内，研究者提出了数千种 {{Term|generative adversarial network|GAN}} 变体，解决了训练不稳定性问题（{{Term|wasserstein gan|WGAN}}、{{Term|spectral normalization|谱归一化}}），实现了条件生成（{{Term|conditional gan|cGAN}}、pix2pix），实现了照片级真实的{{Term|column space|图像}}合成（{{Term|stylegan|StyleGAN}}、{{Term|biggan|BigGAN}}），并扩展到视频、3D 及其他模态。{{Term|adversarial training|对抗训练}}原理还被应用于{{Term|domain adaptation|领域自适应}}、{{Term|data augmentation|数据增强}}、超分辨率以及文本到{{Term|column space|图像}}的生成。 |