Learning path
Generative Models: GANs, VAEs and Diffusion
GANs, autoencoders and VAEs, and how diffusion image generators work, with animated films and runnable Python. Part of the Gen AI Developer premium track.
-
1
GANs: A Generator and a Critic
How Generative Adversarial Networks create realistic images: a forger network versus a detective network, the adversarial training loop, DCGAN, StyleGAN, CycleGAN, and a 1-D GAN-style game you can…
-
2
Autoencoders and Variational Autoencoders (VAEs)
Compress data into a small code and rebuild it: autoencoders for compression and anomaly detection, VAEs for generation with a smooth latent space, the reparameterisation trick, and VAEs…
-
3
Diffusion Models: How Image Generators Work
The idea behind Stable Diffusion, DALL·E-style and Imagen-style generators: add noise step by step, train a network to remove it, then generate from pure noise, guided by text.…