Akademik Veri Yönetim Sistemi
Avrupa Bilim ve Teknoloji Dergisi, cilt.0, sa.Ejosat Özel Sayı 2020 (ARACONF), ss.585-590, 2020 (Hakemli Dergi)
Generative Adversarial Networks (GAN) which are analyzed in this study are among many deep learning methods which have beendeveloped to overcome the restrictions of generic deep learning algorithms such as Restricted Boltzmann Machines (RBM), DeepBoltzmann Machines (DBM) and Variational Autoencoders (VAE). GAN models and structures can create new unique data from thecollected data bases. These data bases can contain thousands of data and different types of data. The variations of these methods aremostly used for deep learning applications such as image restoration and creation, signal processing, and detection of cyber-attacks.In the literature, there are many different types of GANs. In this paper, it was focused on two GAN methods which are the Least Squares Generative Adversarial Networks (LSGAN), and Deep Convolutional Generative Adversarial Networks (DCGAN). Thesemethods have been developed to improve the performance of the traditional GAN algorithm and solve various problems by satisfyingdifferent requirements. In this study, the architectures, usage types, properties and numeric definitions about these two methods weregiven and also the differences between them were analyzed. After that, the practical applications of these algorithms in the literaturewhich have been used for creating new and unique data from the collected data were also discussed in this paper. 5 literature studiesfor LSGAN and 2 literature studies for DCGAN were given. Finally, we have compared the obtained results of these methods andexplain which method can be used for which type of application. As seen from the researches, the applications that these methods canbe applied are different from each other.