A Brief Overview on Generative Adversarial Networks

  • Raj PatelEmail author
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 847)


Generative models dwell into a research domain, which incorporates the use of mathematical models by which machines could mimic and generate new data. Generative models are used in tasks such as image or music generation as well as in capturing and mimicking certain features of a dataset. One such generative model is generative adversarial network, which makes use of a generator and discriminator in order to perform such generative tasks. It does this by maintaining a min–max relationship between the two models. This study gives a brief overview of generative adversarial networks its optimization function and metrics as well as its implementation and working. The study also incorporates a case study, which provides a clear illustration of the actual training, which is to be performed. The performance and the results are sampled and very well compared. The study also gives insights regarding the various applications and the variations of generative adversarial network.


Deep learning Generative adversarial networks GAN Neural networks 


  1. 1.
    Goodfellow, I.J., Pouget-Abadie, J., Warde-Farley, D.: Generative Adversarial Networks, pp. 1–6. Arxiv (2014)Google Scholar
  2. 2.
    Goodfellow, I.: NIPS 2016 Tutorial: Generative Adversarial Networks, pp. 6–13. Arxiv (2016)Google Scholar
  3. 3.
    Welling, M., Kingma, D.P.: Auto-Encoding Variational Bayes, pp. 1–8. Arxiv (2013)Google Scholar
  4. 4.
    Gan, Z., Henao, R., Yuan, X., Li, C., Stevens, A., Carin, L., Pu, Y.: Variational autoencoder for deep learning of images, labels and captions. In: NIPS (2016)Google Scholar
  5. 5.
    Hinton, G., Salakhutdinov, R.: Deep Boltzmann Machines. MLR (2009)Google Scholar
  6. 6.
    Karazeev, A.: Generative Adversarial Networks (GANs): Engine and Applications, 17-8-2017. Available: (Online)
  7. 7.
    Luo, P., Wang, X., Tang, X., Liu, Z.: Deep learning face attributes in the wild. In: Proceedings of International Conference on Computer Vision (ICCV) (2015)Google Scholar
  8. 8.
    Metz, L., Chintala, S., Radford, A.: Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks, pp. 1–7. Arxiv (2015)Google Scholar
  9. 9.
    Chintala, S., Bottou, L., Arjovsky, M.: Wasserstein GAN, pp. 1–5. Arxiv (2017)Google Scholar
  10. 10.
    Osindero, S., Mirza, M.: Conditional Generative Adversarial Nets, pp. 1–5. Arxiv (2014)Google Scholar
  11. 11.
    Duan, Y., Houthooft, R., Schulman, J., Sutskever, I., Abbeel, P., Chen, X.: InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial NetsInfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets, pp. 1–12. Arxiv (2016)Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2019

Authors and Affiliations

  1. 1.Department of Information TechnologyDwarkadas J. Sanghvi College of EngineeringMumbaiIndia

Personalised recommendations