Variational Autoencoders (VAE)

02 Nov 2024

🚧 Work in progress…

This article will cover Variational Autoencoders (VAEs), a powerful class of generative models that combine deep learning with variational inference.

Topics to cover:

• Introduction to autoencoders
• The generative modeling problem
• Variational inference framework for VAEs
• The reparameterization trick
• Evidence Lower Bound (ELBO) in VAEs
• Encoder and decoder architectures
• Training VAEs
• Applications: image generation, representation learning
• Variants: β-VAE, Conditional VAE, Hierarchical VAE
• Comparison with other generative models (GANs, diffusion models)