□ 주제 : Generative Adversarial Networks: From Foundation to Applications
□ 연사 : 심현정 교수 (연세대학교)
□ 시간 : 12/1(수) 15:00 ~ 17:00
□ 온라인 링크 : https://us02web.zoom.us/j/88579691003?pwd=bVdrNHY5djdWeG1zbnZDVzViYktuZz09
□ 초록
In this talk, I will introduce the idea of generative adversarial networks, interesting research topics and their recent process. Despite the impressive success in image generation tasks, GANs are also notorious for several drawbacks, such as mode collapse and training instability. I will briefly cover several seminal works in GAN's family, standard techniques for mitigating those limitations. Finally, the popular downstream task, image-to-image translation using GANs, will be presented to show the potential of GANs in practical applications