Using a Conditional Generative Adversarial Network to Control the Statistical Characteristics of Generated Images for IACT Data Analysis by Julia Dubenskaya et al. on Wednesday 30 NovemberGenerative adversarial networks are a promising tool for image generation inthe astronomy domain. Of particular interest are conditional generativeadversarial networks (cGANs), which allow you to divide images into severalclasses according to the value of some property of the image, and then specifythe required class when generating new images. In the case of images fromImaging Atmospheric Cherenkov Telescopes (IACTs), an important property is thetotal brightness of all image pixels (image size), which is in directcorrelation with the energy of primary particles. We used a cGAN technique togenerate images similar to whose obtained in the TAIGA-IACT experiment. As atraining set, we used a set of two-dimensional images generated using the TAIGAMonte Carlo simulation software. We artificiallly divided the training set into10 classes, sorting images by size and defining the boundaries of the classesso that the same number of images fall into each class. These classes were usedwhile training our network. The paper shows that for each class, the sizedistribution of the generated images is close to normal with the mean valuelocated approximately in the middle of the corresponding class. We also showthat for the generated images, the total image size distribution obtained bysumming the distributions over all classes is close to the originaldistribution of the training set. The results obtained will be useful for moreaccurate generation of realistic synthetic images similar to the ones taken byIACTs.arXiv: http://arxiv.org/abs/http://arxiv.org/abs/2211.15807v1