In this seminar, Generative Adversarial Networks (GANs), will be introduced. The seminar will begin with a simple explanation of the basic working principles of deep learning and how artificial neural networks learn. Then, Convolutional Neural Networks (CNNs), which are widely used in image processing, will be briefly explained.

After that, the architecture of GANs will be discussed, focusing on how the generator and discriminator networks compete with each other to produce realistic data. Important GAN variants such as DCGAN, Conditional GAN, CycleGAN, and StyleGAN will be introduced along with their real world applications. Finally, the challenges, limitations, and ethical considerations related to GANs will be addressed.