Deep Learning Generative Models 3

Question 1

What is the ELBO (Evidence Lower BOund) in variational inference used in VAEs?

Accepted Answer

A lower bound on the log-likelihood of data, maximized to train the VAE by balancing reconstruction and KL terms

Answer

The ELBO = E[log p(x|z)] − KL(q(z|x) || p(z)) provides a tractable lower bound on log p(x), decomposing into reconstruction quality and latent space regularization.

Question 2

What is the Fréchet Inception Distance (FID) used to measure?

Accepted Answer

The quality and diversity of generated images by comparing statistics of generated and real image feature distributions

Answer

FID measures the Fréchet distance between Inception feature distributions of real and generated images, with lower values indicating more realistic and diverse generated samples.

Question 3

What is a denoising score matching objective used to train diffusion models?

Accepted Answer

Training a neural network to predict the noise added at each diffusion step, enabling iterative denoising

Answer

Denoising score matching trains the network to predict the noise ε added to a clean image at each noise level, allowing iterative reconstruction by subtracting predicted noise.

Question 4

What is a conditional diffusion model like Stable Diffusion conditioned on?

Accepted Answer

Text embeddings from a language model like CLIP or a text encoder, provided via cross-attention during denoising

Answer

Conditional diffusion models inject text conditioning via cross-attention between noisy image features and text embeddings at each denoising step, guiding generation toward the prompt.

Question 5

What distinguishes a generative model from a discriminative model?

Accepted Answer

Generative models learn the joint distribution p(x,y) and can generate samples; discriminative models learn p(y|x) for classification only

Answer

Generative models model the full data distribution and can synthesize new samples, while discriminative models learn the decision boundary between classes for prediction tasks only.

Deep Learning Practice Test

Deep Learning Practice Test

Deep Learning Generative Models 3