GAN
1. List
List of GAN Implementation: wiseodd, 설명 BLog
2. Paper
- Learning from Simulated and Unsupervised Images through Adversarial Training : Ashish Shrivastava, Apple, 216
- the blueprint for training state-of-the-art neural nets from only synthetic and unlabelled data
- 영문 정리 글 : SimGANs - a game changer in unsupervised learning, self driving cars, and more
- Improved Techniques for Training GANs: 논문
3. Article (Post, blog, etc.)
지적 대화를 위한 깊고 넓은 딥러닝 (Feat. TensorFlow): [추천]: youtube, 김태훈, PyCon APAC 2016
GAN by Example using Keras on Tensorflow Backend: Rowel Atienza, 코드포함
On the intuition behind deep learning & GANs — towards a fundamental understanding
Generative Adversarial Networks (GANs) in 50 lines of code (PyTorch)
Fantastic GANs and where to find them: GAN의 여러 모형들 설명
GAN 그리고 Unsupervised Learning: 개요로 읽기 적당한글, t-robotics블로그, 테크M 투고글InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets: 논문, ppt정리
아주 간단한 GAN구현하기: 참고 GitHub, Generative Adversarial Network for approximating a 1D Gaussian distribution
NIPS 2016 Tutorial:Generative Adversarial Networks: 논문,youtube, 논문 및 저자 설명
[초짜 대학원생 입장에서 이해하는 시리즈]
Domain-Adversarial Training of Neural Networks (DANN) : #1, #2, #3
Deep Convolutional Generative Adversarial Network (DCGAN): #1, #2
3. Tutorial (Series, )
4. Youtube
6. Material (Pdf, ppt)
- Generative Adversarial Network and its Variations: ppt, GAN 변형들
7. Implementation (Project)
Keras Adversarial Models: bstriner
13줄 간단 코드 및 설명: nlintz
SimGan: Keras 코드
An introduction to Generative Adversarial Networks (with code in TensorFlow): 원문,
정리, GitHubLeast Squares Generative Adversarial Networks(LSGAN), 구현시 어려웠던점
SimGAN_NYU_Hand : Simulated+Unsupervised (S+U) learning in TensorFlow /w NYU Hand Dataset
Generative Adversarial Networks (GANs) in 50 lines of code (PyTorch): Dev Nag's 미디엄 포스트
DiscoGAN: SKTBrain, Pythrch기반
- DiscoGAN 설명자료
- DiscoGAN Arxiv 논문 링크
- Taehoon Kim 님이 구현하신 DiscoGAN 소스코드
- wiseodd 님이 구현하신 DiscoGAN 소스코드
List of generative models: 거의 모든 GAN코드 모음
DCGAN-MNIST: Keras버젼, erilyth