Generative Adversarial Networks (GANs): A Comprehensive List of Influential Papers

Generative Adversarial Networks (GANs): Key Papers and Their Impact

This list compiles influential papers that have shaped the field of Generative Adversarial Networks (GANs), a powerful deep learning technique for generating realistic data.

1. 'Generative Adversarial Networks' by Ian Goodfellow et al., published in NIPS 2014 (Conference paper) This foundational paper introduced the concept of GANs, laying the groundwork for subsequent research.

2. 'Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks' by Alec Radford et al., published in arXiv 2015 (Preprint) This work explored the application of GANs for unsupervised representation learning, demonstrating their ability to learn meaningful features from data.

3. 'Improved Techniques for Training GANs' by Tim Salimans et al., published in NIPS 2016 (Conference paper) This paper introduced techniques to improve the stability and performance of GAN training, addressing key challenges in the field.

4. 'Conditional Generative Adversarial Nets' by Mehdi Mirza and Simon Osindero, published in arXiv 2014 (Preprint) This work extended GANs to conditional generation, enabling the control of the generated output based on input conditions.

5. 'Progressive Growing of GANs for Improved Quality, Stability, and Variation' by Tero Karras et al., published in ICLR 2018 (Conference paper) This paper introduced a novel approach to GAN training, progressively growing the generator and discriminator networks, leading to higher-quality and more diverse outputs.

6. 'CycleGAN: Unpaired Image-to-Image Translation Using Cycle-Consistent Adversarial Networks' by Jun-Yan Zhu et al., published in ICCV 2017 (Conference paper) CycleGAN demonstrated the ability of GANs to translate images between different domains without requiring paired data.

7. 'A Style-Based Generator Architecture for Generative Adversarial Networks' by Tero Karras et al., published in CVPR 2019 (Conference paper) This paper introduced a style-based generator architecture, leading to more control over the generated image style and improved image quality.

8. 'BigGAN: Large Scale GAN Training for High Fidelity Natural Image Synthesis' by Andrew Brock et al., published in arXiv 2018 (Preprint) BigGAN pushed the boundaries of GANs by training on large datasets and achieving high-fidelity image synthesis.

9. 'StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translation' by Yunjey Choi et al., published in CVPR 2018 (Conference paper) StarGAN unified image translation across multiple domains using a single GAN, enabling efficient and versatile image manipulation.

10. 'Generative Adversarial Networks for Image and Video Synthesis: Algorithms and Applications' by Ting-Chun Wang et al., published in arXiv 2019 (Preprint) This comprehensive review paper covered various GAN algorithms and their applications in image and video synthesis.

Sources:

Most of the listed papers are high-level conference papers, with a few being preprints available on arXiv.

This list highlights some of the most significant papers in the field of GANs, demonstrating their evolution and impact on various applications, including image synthesis, image translation, and representation learning.