Course positioning
CS 236G is a special-topics course on generative models focused on GANs. It goes from basic adversarial training to SOTA variants such as StyleGAN, discusses mode collapse, training stability, evaluation metrics, and bias, and covers applications in image editing, medical low-data settings, and climate visualization.
What it covers
- GAN basics: generator/discriminator, minimax, losses
- Training techniques: WGAN, spectral normalization, TTUR, regularization
- Conditional generation: cGAN, StyleGAN, image-to-image
- Evaluation: FID, IS, precision/recall
- Bias and ethics: data bias, deepfakes
- Applications: editing, medicine, scientific visualization
- Project
Prerequisites
Linear algebra, statistics, CS 106B; a graduate AI course (CS 229/230/221).
Assessment & workload
Assignments plus a project. Check the current syllabus for details.
Who it's for
Image generation, visual editing, and generative-model engineering; people who want to understand GANs rather than only calling diffusion.
Study advice
Take CS 236 or CS 231N first. Reproduce a GAN training-stability failure once by hand.
Related courses
Official course description
Generative Adversarial Networks (GANs) have rapidly emerged as the state-of-the-art technique in realistic image generation. This course presents theoretical intuition and practical knowledge on GANs, from their simplest to their state-of-the-art forms. Their benefits and applications span realistic image editing that is omnipresent in popular app filters, enabling tumor classification under low data schemes in medicine, and visualizing realistic scenarios of climate change destruction. This course also examines key challenges of GANs today, including reliable evaluation, inherent biases, and training stability. After this course, students should be familiar with GANs and the broader generative models and machine learning contexts in which these models are situated. Prerequisites: linear algebra, statistics, CS106B, plus a graduate-level AI course such as: CS230, CS229 (or CS129), or CS221.