什么是GAN？
GAN（Generative Adversarial Netwo，生成对抗网络）是用于无监督学习的机器学习模型，由Ian Goodfellow等人在2014年提出，由神经网络构成判别器和生成器构成，通过一种互相竞争的机制组成的一种学习框架。
卷积神经网络之父-Yann LeCun这样评论GAN：
在我看来，最重要的是对抗训练( GAN也称为生成对抗网络)。这一想法最初是Ian Goodfellow在蒙特利尔大学读书是提出的，他当时是Yoshua Bengio的学生(Yoshua Bengio先加入了Google Brain，最近有离职加入OpenAI )。在我看来，这一想法与正在被提出的各种变化，是最近十年来在ML中最有趣的想法。
主要整理了自2014年，GAN推出以来，一些优质的论文，分享给有需要的朋友。
（限于篇幅原因，本文主要列出前50篇GAN相关论文，文末附上完整且带论文链接的list）
列表如下：
1. 3C-GAN: AN CONDITION-CONTEXT-COMPOSITE GENERATIVE ADVERSARIAL NETWORKS FOR GENERATING IMAGES SEPARATELY
2018/1 ICLR2018 3C-GAN
2. 3D conditional generative adversarial networks for high-quality PET image estimation at low dose
2018/7 Medical: Reconstruction New
3. 3D Consistent Biventricular Myocardial Segmentation Using Deep Learning for Mesh Generation
2018/3 New, 3D
4. 3D Medical Image Synthesis using Generative Adversarial Networks
2017/ Medical: Synthersize Medical
5. 3D Object Reconstruction from a Single Depth View with Adversarial Learning?
2017/8 Applied Vision 3D-RecGAN
6. 3D Reconstruction of Incomplete Archaeological Objects Using a Generative Adversary Network
2017/11 ORGAN
8. 3D Shape Induction from 2D Views of Multiple Objects
2016/12 3D Object generation PrGAN Citation: 9
9. 3D-Scene-GAN: Three-dimensional Scene Reconstruction with Generative Adversarial Networks
2018/2 ICLR2018
10. A Classification-Based Perspective on GAN Distributions
2018/1 ICLR2018
11. A Classification-Based Perspective on GAN Distributions?
2017/11 Theory & Machine Learning Citation: 1
12. A conditional adversarial network for semantic segmentation of brain tumor
2018/2 Medical: Segmentation New
13. A Connection between Generative Adversarial Networks, Inverse Reinforcement Learning, and Energy-Based Models?
2016/11 Theory & Machine Learning Citation: 16
14. A Deep Generative Adversarial Architecture for Network-Wide Spatial-Temporal Traffic State Estimation
2018/1 None
15. A Deep Predictive Coding Network for Learning Latent Representations
2018/3 New, Bio
16. A General Retraining Framework for Scalable Adversarial Classification?
2016/4 Theory & Machine Learning Citation: 6
17. A Generalized Active Learning Approach for Unsupervised Anomaly Detection
2018/5 New
18. A generative adversarial framework for positive-unlabeled classification
2017/11 GPU
19. A Generative Model for Volume Rendering
2017/10 Medical: Volume Rendering Applied Other
20. A Hybrid Model for Identity Obfuscation by Face Replacement
2018/4 New
21. A Multi-Discriminator CycleGAN for Unsupervised Non-Parallel Speech Domain Adaptation
2018/4 New
22. A Novel Approach to Artistic Textual Visualization via GAN?
2017/10 Applied Vision GAN-ATV
23. A Self-Training Method for Semi-Supervised GANs
2018/1 ICLR2018
24. A Solvable High-Dimensional Model of GAN
2018/5 New
25. A step towards procedural terrain generation with GANs?
2017/7 Applied Vision
26. A Study into the similarity in generator and discriminator in GAN architecture
2018/2 None
27. A Study of Cross-domain Generative Models applied to Cartoon Series?
2017/9
28. A survey of image synthesis and editing with generative adversarial networks
2017/12 Medical: Synthersize New
29. A Variational Inequality Perspective on Generative Adversarial Nets
2018/2 None
30. A-Fast-RCNN: Hard Positive Generation via Adversary for Object Detection?
2017/4 Object Detection Citation: 12
31. ABC-GAN: Adaptive Blur and Control for improved training stability of Generative Adversarial Networks
2017/8 ABC-GAN Stars: 2
32. Abnormal Event Detection in Videos using Generative Adversarial Nets?
2017/8 Applied Vision
33. Accelerated Magnetic Resonance Imaging by Adversarial Neural Network
2017/9 Medical: Reconstruction
34. Accelerating Science with?Generative?Adversarial?Networks: An Application to 3D Particle Showers in Multilayer Calorimeters
2018/1 Medical: Reconstruction New
35. Activation Maximization Generative Adversarial Nets
2017/3 Theory & Machine Learning AM-GAN
36. Activation Maximization Generative Adversarial Nets
2018/1 ICLR2018
37. ACtuAL: Actor-Critic Under Adversarial Learning
2017/11 ACtuAL
38. AdaGAN: Boosting Generative Models?
2017/1 Ensemble, Theory & Machine Learning AdaGAN Citation: 19
39. Adaptive template generation for amyloid PET using a deep learning approach
2018/5 Medical: Synthersize New
40. Addressing Challenging Place Recognition Tasks using Generative Adversarial Networks?
2017/9
41. Adversarial Attacks on Neural Network Policies?
2017/2 Citation: 21
42. Adversarial Autoencoders?
2015/11 Theory & Machine Learning AAE Citation: 163 Stars: 130
43. Adversarial Data Programming: Using GANs to Relax the Bottleneck of Curated Labeled Data
2018/3 New
44. Adversarial Deep Structural Networks for Mammographic Mass Segmentation
2016/12 BioarXiv
45. Adversarial Deep Structural Networks for Mammographic Mass Segmentation?
2016/12 Semantic Segmentation Citation: 7
46. Adversarial Deep Structured Nets for Mass Segmentation from Mammograms
2017/10 Medical: Segmentation Stars: 13
47. Adversarial Discriminative Domain Adaptation?
2017/2 Theory & Machine Learning Citation: 52
48. Adversarial examples for generative models
2017/2 Adversarial Examples (Defense vs Attack)
49. Adversarial Examples for Semantic Segmentation and Object Detection?
2017/7 Citation: 17
50. Adversarial Examples Generation and Defense Based on Generative Adversarial Network
2017/ Adversarial Examples (Defense vs Attack)
