Research on Generative Adversarial Networks

Yuanyuan Zhang ； Yongli Zhu

《International Core Journal of Engineering》 8卷2期 (2022/02) Pp. 265-269

https://doi.org/10.6919/ICJE.202202_8(2).0033

摘要

Regarding the hotspot model of GAN, first introduces the research status of GAN; then introduces the principle of GAN and the problems in the training process of GAN; then discusses the derivative model of GAN from the improvement of network structure and loss function, and its typical The improvement points, applicable scenarios, advantages and disadvantages of GAN model, and finally summarizes the application of GAN and future research directions.

關鍵字

Generative Adversarial Network (GAN) ； Gradient Disappears ； Mode Collapse

參考文獻

Zhang H,Xu T,Li H S,et al.Stackgan:text to photo-realistic image synthesis with stacked generative adversarial networks[C]//Proceedings of the 2017 IEEE International Conference on Computer Vision,Venice,Oct22-29,2017.Washington:IEEEComputerSociety,2017:5908-5916.

Zhu J Y,Pakr T,Isola P,et al.Unpaired image-to-image translation using cycle consistent adversarial networks[C]//Proceedings of the 2017 IEEE International Conference on Computer Vision, Venice, Oct22-29, 2017.Washington:IEEE Computer Society,2017:2242-2251.

Mao X D,Li Q,Xie H R,et al.Least squares generative adversarial networks[C]//Proceedings of the 2017 IEEE International Conference on Computer Vision, Venice, Oct22-29, 2017. Washington: IEEE Computer Society, 2017:2813-2821.

Goodfellow I J,Pouget-AbadieJ, MirzaM, etal. Generative adversarialnets[C]// Proceedings of the 27th International Conference on Neural Information Processing Systems, Montreal, Dec8-13, 2014. Cambridge: MITPress,2014:2672-2680.

Goodfellow I. NIPS 2016 tutorial: generative adversarial networks[J]. arXiv preprint arXiv:1701.00160, 2016.