> ML_LITERATURE // GOODFELLOW-2014-GENERATIVE-ADVERSARIAL-NETS_v1.0
Generative Adversarial Nets (GAN)
Ian J. Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, Yoshua Bengio · Advances in Neural Information Processing Systems (NeurIPS) (2014)
foundational2014foundationalthirdPartyReproduced
Principal Contribution
Framed generative modeling as a two-player zero-sum minimax game between a generator and a discriminator without Markov chains or approximate inference.
Operational Relevance
Pioneered realistic synthetic image generation, deepfakes, neural style transfer, and discriminator loss objectives across computer vision.
Assumptions
- Minimax optimization converges to the Nash equilibrium where the generator recovers the true data distribution and discriminator probability is 1/2
Limitations
- Training instability, vanishing gradients under saturated discriminators, and catastrophic mode collapse
Connected Algorithms, Architectures & Tools
Related Algorithms:
Related Architectures:
Implementing Libraries: