Methoden vergleichen

	Generative Adversarial Network	Neuronale Stilübertragung	Wasserstein GAN (WGAN)
Fachgebiet	Deep Learning	Deep Learning	Deep Learning
Familie	Machine learning	Machine learning	Machine learning
Entstehungsjahr	2014	2015	2017
Urheber	Goodfellow, I. et al.	Gatys, L. A.; Ecker, A. S.; Bethge, M.	Martín Arjovsky, Soumith Chintala & Léon Bottou
Typ	Generative deep learning (adversarial two-network game)	Iterative optimization over CNN feature statistics	Generative adversarial network variant
Wegweisende Quelle	Goodfellow, I. et al. (2014). Generative Adversarial Nets. NeurIPS.	Gatys, L. A., Ecker, A. S., & Bethge, M. (2016). Image Style Transfer Using Convolutional Neural Networks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 2414–2423.	Arjovsky, M., Chintala, S., & Bottou, L. (2017). Wasserstein generative adversarial networks. International Conference on Machine Learning (ICML), 214–223.
Aliasnamen	Üretici Çekişmeli Ağ (GAN), GAN, generative adversarial nets, adversarial network	NST, artistic style transfer, neural artistic style, CNN style transfer	WGAN, Earth-Mover GAN, Wasserstein Generative Adversarial Network, Wasserstein-GAN
Verwandt	4	3	3
Zusammenfassung	A Generative Adversarial Network (GAN), introduced by Ian Goodfellow and colleagues in 2014, produces realistic synthetic data through the competition of two neural networks — a generator and a discriminator. It is widely used for image synthesis, data augmentation, and distribution estimation.	Neural Style Transfer (NST) is a deep-learning image synthesis technique, introduced by Gatys, Ecker, and Bethge in 2015, that separates the semantic content of one image from the visual texture and artistic style of another, then recombines them into a single synthesized image by iteratively optimizing pixel values to minimize a combined content and style loss computed from the feature maps of a pretrained convolutional neural network.	Wasserstein GAN (WGAN) is a generative adversarial network variant introduced by Arjovsky, Chintala, and Bottou in 2017 that replaces the Jensen-Shannon divergence used in the original GAN with the Wasserstein-1 (Earth Mover) distance. This substitution provides a theoretically grounded training objective that yields more stable optimization and a loss value that correlates meaningfully with generated sample quality, addressing the notorious mode collapse and vanishing gradient problems of standard GANs.