Topic 09: Generative Adversarial Neural Networks
Generative Adversarial Networks (GANs) are a class of machine learning models that consist of two competing networks: a generator, which creates data samples, and a discriminator, which evaluates their authenticity. This chapter introduces the core principles of GANs, explores popular variants and addresses challenges in optimizing them.
-
Chapter 09.01: Generative Adversarial Networks (GANs)
This subchapter introduces GANs, which consist of a generator and a discriminator competing to create realistic data samples. In addition, key concepts discussed in this subchapter include the minimax loss function and challenges in training, such as stability issues and optimal discriminator requirements.
-
Chapter 09.02: GAN variants
This subsection introduces key GAN variants that address limitations in traditional GAN training. It covers non-saturating loss, which helps avoid vanishing gradients. Conditional GANs are also discussed, where additional data, such as labels or images, is used to guide the generation process, enabling more controlled and targeted outputs.
-
Chapter 09.03: Challenges for GAN Optimization
This subsection covers the main challenges in optimizing GANs, including non-convergence, mode collapse, and oscillating or chaotic behaviors during training.