Generative Models

From latent variables, variational inference, and VAEs to diffusion, continuous flows, and flow matching, with image and geometry examples.

Collection contents 11 articles

  1. From Latent Variable Models to Variational Inference
  2. Neural Networks as Probability Models
  3. Trainable Variational Inference: Sampling, Reparameterization, and Amortization
  4. VAE: From a Latent Probability Model to Generation
  5. Diffusion Models I: From Corruption and Denoising to Generation
  6. Diffusion Models II: A Compact Derivation from Corruption to Sampling
  7. Diffusion Models III: DDPM and DDIM Share Training, Not Sampling Rules
  8. Continuous-Time Generation I: One Score, SDE or ODE Sampling
  9. Continuous-Time Generation II: Invertible Maps and Normalizing Flows
  10. Flow Matching: Learning Motion from Noise to a Chair
  11. Geometry Generation: ShapeNet Chairs as Point Clouds or Fields

The first four articles cover probability modeling and VAEs. Start diffusion at article five: the first article explains corruption, training, generation, and common extensions using MNIST and ShapeNet chairs. The second gives the complete DDPM mathematical chain with a compact derivation. Continue to faster sampling, continuous time, Flow Matching, and SDF geometry.

Back to articles