Extra-Tropical Cyclones (ETCs) are major storm system ruling and influencing the atmospheric structure at mid-latitudes. These events are usually characterized by strong winds and heavy precipitation and cause considerable storm surges with threatening wave systems for coastal regions. The possibility to simulate these storms or to increase the amount of significant data available is crucial to optimize risk assessment and risk management for construction projects and territorial plans which might get damaged by events of this kind. The project addresses the possibility to learn the distribution of cyclones atmospheric fields of pressure, wind and precipitation in the North Atlantic by training a Generative Adversarial Network (GAN). The ETCs tracks are extracted from the ERA5 reanalysis dataset in the domain with boundaries 0°-90°N, 70°W-20°E and period going from 1st January 1979 to 1st January 2020. A GAN tries to learn the distribution of a training set based on a game theoretic scenario where two network competes against each other, the generator and the discriminator. The former is trained to generate new examples which are plausible and similar to the real ones by having as input a vector of random Gaussian values. The random vectors domain is called latent space. The latter learns to distinguish whether an example is coming from the dataset distribution or not. The competition set by the game scenario makes the network improve until the counterfeits are indistinguishable form the original. The generative models trained on the ETCs dataset are validated to understand if they are able to generate new samples of fields presenting similar atmospheric characteristics to those of the original dataset. To train the GAN two different loss function are considered, the Wasserstein distance and the Cramèr distance. The Cramèr Gan (CGAN) shows better performance in representing the distribution of the atmospheric fields, generating images that on average look similar to the original ones. The Wasserstein GAN (WGAN) behaviour shows poor performance in representing the precipitation in general, but it is able to similarly reproduce the values distribution for what concerns pressure and wind. The images generated by the WGAN have many differences compared to the original ones and are very blurry with particular data structures that looks like artefacts built by the network. The atmospheric structure of the images generated by the CGAN is investigated by considering 4 cyclones as case study and comparing the frames of their tracks to those of synthetic tracks generated by linear interpolation. The linear interpolation is performed between the random vectors generating the most similar images to the initial and final snapshot of the original track. The interpolated images show interesting features in the similarity with the original track, which suggest that the network has learned a representation of the ETCs fields that is promising for further investigation.