Phase 1: Convolutional Neural Networks · ~75 minutes · Python

Image Generation — GANs

A GAN is two neural networks in a fixed game. One draws, one critiques. They get better together until the drawings fool the critic.

What you will learn

Explain the minimax game between generator and discriminator and why the equilibrium corresponds to p_model = p_data
Implement a DCGAN in PyTorch and get it to generate coherent 32x32 synthetic images in under 60 lines
Stabilise GAN training with the three standard tricks: non-saturating loss, spectral norm, TTUR (two-timescale update rule)
Read training curves that distinguish healthy convergence from mode collapse, oscillation, and discriminator-wins-completely

Introduction

Type: Build
Languages: Python
Prerequisites: the Deep Learning from Scratch course, Optimizers, Regularization
Time: ~75 minutes

Objective

Learning objectives

Explain the minimax game between generator and discriminator and why the equilibrium corresponds to p_model = p_data
Implement a DCGAN in PyTorch and get it to generate coherent 32x32 synthetic images in under 60 lines
Stabilise GAN training with the three standard tricks: non-saturating loss, spectral norm, TTUR (two-timescale update rule)
Read training curves that distinguish healthy convergence from mode collapse, oscillation, and discriminator-wins-completely

Unlock the full lesson

You've read the first 2 sections. The rest of this lesson covers The Problem, The Concept, Build, Check Yourself, Key Terms & Next — plus a hands-on lab, quiz, and project artifact.

Create a free account — the opening phases of 29 of 35 courses are free, no credit card.