Neural networks have helped make machine learning considerably more efficient. Seeing the use of various constructs of neural networks being quite direct, of course with necessary tuning, it can said that over a while they have also got a notion of templates. Generative Adversarial Networks (GANs) too, is considered as a template of neural networks. It involves the use of constant feedback and feedforward loops kind of arrangement of networks. This cycle progressively improves output. The emergence of GANs saw the unfolding of possibilities to solve many problems that one couldn't at all or even if one could solve them, the solutions were not feasible. Its potential is seen on a huge scale. It can also bring about impact in many creative areas like fashion, entertainment, etc. Because of the very construct of GANs, it helps make a lot of problems that were in the creative space, now doable. In this paper, the authors have discussed the initial points of influence that directly or even indirectly influenced the need and emergence of GANs, but before diving into that they have articulated a brief overview of the working of GANs and finally have put forth a classification scheme for the applications of GANs based on currently seen uses of it.

Generative and Adversarial Networks (GANs) Neural Network Machine Learning.