生成对抗网络
| 机器学习与数据挖掘 |
|---|
生成对抗网络(英语:Generative Adversarial Network,简称GAN)是非监督式学习的一种方法,通过两个神经网路相互博弈的方式进行学习。该方法由伊恩·古德费洛等人于2014年提出。[1] 生成对抗网络由一个生成网络与一个判别网络组成。生成网络从潜在空间(latent space)中随机取样作为输入,其输出结果需要尽量模仿训练集中的真实样本。判别网络的输入则为真实样本或生成网络的输出,其目的是将生成网络的输出从真实样本中尽可能分辨出来。而生成网络则要尽可能地欺骗判别网络。两个网络相互对抗、不断调整参数,最终目的是使判别网络无法判断生成网络的输出结果是否真实。[1][2][3]

生成对抗网络常用于生成以假乱真的图片。[4]此外,该方法还被用于视频帧预测[5]、三维物体模型[6]等。
生成对抗网络虽然最开始提出是为了无监督学习,但经证明对半监督学习[4]、完全监督学习[7]、强化学习[8]GAIL(Generative Adversarial Imitation Learning)通过逆强化学习框架实现策略优化[9]也有效。 在2016年的一个研讨会上,杨立昆称生成式对抗网络为“机器学习这二十年来最酷的想法”[10]。
核心定义
[编辑]- 数学形式 minGmaxDV(D,G)=Ex∼pdata[logD(x)]+Ez∼pz[log(1−D(G(z)))] 其中G为生成器,D为判别器[11]
- 潜在空间说明,潜在空间z通常服从高斯分布N(0,I),维度需人工设置(如DCGAN中z∈R100)[12]
| 模型 | 数据集 | 评价指标 (FID↓) | 参数量 |
|---|---|---|---|
| StyleGAN2 | FFHQ | 2.84 | 30M |
| BigGAN-deep | ImageNet | 3.45 | 50M |
| VQ-VAE-2 | CelebA-HQ | 5.18 | 13M |
重要子类说明
[编辑]- Wasserstein GAN改进 使用Earth-Mover距离替代JS散度: W(pr,pg)=infγ∈Π(pr,pg)E(x,y)∼γ[∣∣x−y∣∣] 需满足判别器Lipschitz约束[14]
- 渐进式训练策略 ProGAN采用分层训练模式,从低分辨率(4×4)开始逐步加倍分辨率至1024×1024[15]
应用
[编辑]时尚和广告
[编辑]生成对抗网路可用于创建虚构时装模特的照片,无需聘请模特、摄影师、化妆师,也省下工作室和交通的开销[18]。 生成对抗网路可用于时尚广告活动,创建来自不同群体的模特儿,这可能会增加这些群体的人的购买意图[19]。
科学
[编辑]生成对抗网路可以改善天文图像[20],并模拟重力透镜以进行暗物质研究[21][22][23]。
在2019年,生成对抗网路成功地模拟了暗物质在太空中特定方向的分布,并预测将要发生的引力透镜。[24][25]
电子游戏
[编辑]在2018年,生成对抗网路进入了电子游戏改造社区。对旧的电子游戏透过图像训练,以4k或更高分辨率重新创建低分辨率2D纹理,然后对它们进行下取样以适应游戏的原始分辨率(结果类似于抗锯齿的超级取样方法)[26]。通过适当的训练,生成对抗网路提供更清晰、高于原始的2D纹理图像品质,同时完全保留原始的细节、颜色。
参见
[编辑]参考文献
[编辑]- ^ 1.0 1.1 Goodfellow, Ian J.; Pouget-Abadie, Jean; Mirza, Mehdi; Xu, Bing; Warde-Farley, David; Ozair, Sherjil; Courville, Aaron; Bengio, Yoshua. Generative Adversarial Networks. 2014. arXiv:1406.2661
[stat.ML].
- ^ 能根據文字生成圖片的 GAN,深度學習領域的又一新星. [2018-04-15]. (原始内容存档于2018-04-15).
- ^ Andrej Karpathy, Pieter Abbeel, Greg Brockman, Peter Chen, Vicki Cheung, Rocky Duan, Ian Goodfellow, Durk Kingma, Jonathan Ho, Rein Houthooft, Tim Salimans, John Schulman, Ilya Sutskever, And Wojciech Zaremba, Generative Models, OpenAI, [2016-04-07], (原始内容存档于2021-04-22)
- ^ 4.0 4.1 Salimans, Tim; Goodfellow, Ian; Zaremba, Wojciech; Cheung, Vicki; Radford, Alec; Chen, Xi. Improved Techniques for Training GANs. 2016. arXiv:1606.03498
[cs.LG].
- ^ 存档副本. [2017-03-17]. (原始内容存档于2017-03-20).
- ^ 3D Generative Adversarial Network. [2017-03-17]. (原始内容存档于2019-10-27).
- ^ Isola, Phillip; Zhu, Jun-Yan; Zhou, Tinghui; Efros, Alexei. Image-to-Image Translation with Conditional Adversarial Nets. Computer Vision and Pattern Recognition. 2017 [2019-06-18]. (原始内容存档于2020-04-14).
- ^ Ho, Jonathon; Ermon, Stefano. Generative Adversarial Imitation Learning. Advances in Neural Information Processing Systems. [2019-06-18]. (原始内容存档于2019-10-19).
- ^ Ermon, Stefano, Probabilistic Inference by Hashing and Optimization, The MIT Press: 265–288, 2016-12-23 [2025-08-07], ISBN 978-0-262-33793-9
- ^ LeCun, Yann. RL Seminar: The Next Frontier in AI: Unsupervised Learning. [2019-06-18]. (原始内容存档于2020-04-30).
- ^ Parker, Charles Thomas; Taylor, Dorothea; Garrity, George M. Exemplar Abstract for Rhodococcus wratislaviensis (Goodfellow et al. 1995) Goodfellow et al. 2002 emend. Nouioui et al. 2018 and Tsukamurella wratislaviensis Goodfellow et al. 1995.. The NamesforLife Abstracts. 2010-03-16 [2025-08-07].
- ^ Radford, Andrew N. Moving beyond species-specific noise-induced changes in birdsong: a comment on Roca et al.. Behavioral Ecology. 2016, 27 (5) [2025-08-07]. ISSN 1045-2249. doi:10.1093/beheco/arw103.
- ^ Figure 3: Risk of bias summary (Abreu et al., 2017; Afshar et al., 2010; Ai, 2020; Bolasco et al., 2011; Cai et al., 2022; Chen, Zhao & Huang, 2019; Dai & Ma, 2021; Deng, 2011; Dong et al., 2011; Fakhrpour et al., 2020; Fang et al., 2023; Feng et al., 2020; Frih et al., 2017; Hristea et al., 2016; Jeong et al., 2019; Kozlowska et al., 2023; Leng, 2012; Li et al., 2008; Li & Feng, 2020; Liao et al., 2016; Limwannata et al., 2021; Lu, 2022; Martin-Alemañy et al., 2020, 2016, 2022; Sezer et al., 2014; Shi et al., 2021; Su et al., 2022; Sun, Sun & Yang, 2022a; Tabibi et al., 2023; Tan et al., 2015; Tayebi, Ramezani & Kashef, 2018; Vijaya et al., 2019; Wang & Liu, 2021; Wang, 2018; Wang et al., 2019; Wang, 2019; Wang et al., 2023; Wei, 2020; Wen et al., 2022; Wilund et al., 2010; Xu et al., 2022; Xu & Fang, 2016; Yan, Zhao & Peng, 2022; Yang et al., 2021; Yao et al., 2020; Yu & Cao, 2018; Zeng et al., 2020; Zhou, 2020; Zhou et al., 2016; Zhu et al., 2020).. doi.org. [2025-08-07].
- ^ Table 1: Study characteristics (Elmasri et al., 2017; Rastan et al., 2008; Ratnam et al., 2007; Deuling et al., 2008; Veasey et al., 2008; Engelbert et al., 2010; Lucatelli et al., 2017; Gonen, Hakyemez & Erdogan, 2021; Ierardi et al., 2023).. doi.org. [2025-08-07].
- ^ Indonesian Comparative Law Review https://doi.org/10.18196/iclr.2018.11. 2018, 1 (1) [2025-08-07]. ISSN 2655-2353. doi:10.18196/iclr.2018.11. 缺少或
|title=为空 (帮助) - ^ Caesar, Holger, A list of papers on Generative Adversarial (Neural) Networks: nightrome/really-awesome-gan, 2019-03-01 [2019-03-02], (原始内容存档于2020-04-30)
- ^ 生成式AI:缘起、机遇和挑战, 经济观察报, 2023-01-09. [2023-01-24]. (原始内容存档于2023-01-24).
- ^ Wong, Ceecee. The Rise of AI Supermodels. CDO Trends. [2019-06-18]. (原始内容存档于2020-04-16).
- ^ Dietmar, Julia. GANs and Deepfakes Could Revolutionize The Fashion Industry. Forbes. [2019-06-18]. (原始内容存档于2019-09-04).
- ^ Schawinski, Kevin; Zhang, Ce; Zhang, Hantian; Fowler, Lucas; Santhanam, Gokula Krishnan. Generative Adversarial Networks recover features in astrophysical images of galaxies beyond the deconvolution limit. Monthly Notices of the Royal Astronomical Society: Letters. 2017-02-01, 467 (1): L110–L114. Bibcode:2017MNRAS.467L.110S. arXiv:1702.00403
. doi:10.1093/mnrasl/slx008.
- ^ Kincade, Kathy. Researchers Train a Neural Network to Study Dark Matter. R&D Magazine. [2019-06-18]. (原始内容存档于2019-05-15).
- ^ Kincade, Kathy. CosmoGAN: Training a neural network to study dark matter. Phys.org. 2019-05-16 [2019-06-18]. (原始内容存档于2020-04-14).
- ^ Training a neural network to study dark matter. Science Daily. 2019-05-16 [2019-06-18]. (原始内容存档于2020-04-30).
- ^ at 06:13, Katyanna Quach 20 May 2019. Cosmoboffins use neural networks to build dark matter maps the easy way. www.theregister.co.uk. [2019-05-20]. (原始内容存档于2020-04-23) (英语).
- ^ Mustafa, Mustafa; Bard, Deborah; Bhimji, Wahid; Lukić, Zarija; Al-Rfou, Rami; Kratochvil, Jan M. CosmoGAN: creating high-fidelity weak lensing convergence maps using Generative Adversarial Networks. Computational Astrophysics and Cosmology. 2019-05-06, 6 (1): 1. ISSN 2197-7909. doi:10.1186/s40668-019-0029-9.
- ^ Tang, Xiaoou; Qiao, Yu; Loy, Chen Change; Dong, Chao; Liu, Yihao; Gu, Jinjin; Wu, Shixiang; Yu, Ke; Wang, Xintao. ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks. 2018-09-01 [2019-06-18]. (原始内容存档于2019-04-13) (英语).