What are normalizing flows and their trade-off vs GANs / diffusion?
hardAnswer
- Sequence of invertible transformations from a simple base (N(0, I)) to the data distribution, using the change-of-variables formula for exact log-likelihood.
- Advantages: exact density evaluation, invertible sampling.
- Disadvantages: constrained architectures (each layer must be invertible with a tractable Jacobian) → less expressive per-parameter than GAN / diffusion.
- Used in density estimation and RL where exact log p is needed.
- Modern generative modeling has mostly shifted to diffusion for quality.
Check yourself — multiple choice
- Not invertible
- Invertible transformations with tractable Jacobian → exact log-likelihood; architectural constraints reduce expressiveness vs diffusion
- Same as GAN
- Only used in NLP
Normalizing flows: invertible + tractable Jacobian → exact density; less expressive than diffusion.
#generative#architectures
Practise Deep Learning
214 interview questions in this topic.
Related questions
- What are the main ideas of StyleGAN?
- How does Latent Diffusion (Stable Diffusion) reduce compute?
- Why do residual (skip) connections help training deep networks?
- Depth vs width — which do you scale first?
- How do you compute the parameter count of a linear layer and a convolution?
- How do FLOPs scale for a forward pass through an MLP and a transformer?