Quiz: Foundations of Deep Learning
Questions of Chapter 2: Understanding Generative Models
Question 6: Generative vs. Discriminative Models
What is the primary difference between generative and discriminative models?
A) Generative models classify data, discriminative models generate new data
B) Generative models learn \(P(Y|X)\), discriminative models learn \(P(X, Y)\)
C) Generative models learn \(P(X, Y)\), discriminative models learn \(P(Y|X)\)
D) Generative models use labeled data, discriminative models use unlabeled data
Question 7: GAN Architecture
What are the two main components of a Generative Adversarial Network (GAN)?
A) Encoder and Decoder
B) Generator and Discriminator
C) Generator and Encoder
D) Discriminator and Decoder
Question 8: Variational Autoencoders (VAEs)
In a Variational Autoencoder (VAE), what is the role of the encoder and decoder?
A) The encoder generates data, the decoder evaluates it
B) The encoder maps input data to a latent space, the decoder generates new data from the latent space
C) The encoder classifies data, the decoder reconstructs it
D) The encoder reduces data dimensionality, the decoder increases it
Question 9: Autoregressive Models
How do autoregressive models generate data?
A) By generating all data points at once
B) By generating one data point at a time, conditioned on the previous points
C) By transforming latent variables
D) By using adversarial training
Question 10: Normalizing Flows
What is a key characteristic of Normalizing Flows?
A) They use a series of non-invertible transformations
B) They provide exact likelihood estimation and efficient sampling
C) They are primarily used for classification tasks
D) They do not require training data
Questions of Chapter 2: Understanding Generative Models
Question 6: Generative vs. Discriminative Models
What is the primary difference between generative and discriminative models?
A) Generative models classify data, discriminative models generate new data
B) Generative models learn \(P(Y|X)\), discriminative models learn \(P(X, Y)\)
C) Generative models learn \(P(X, Y)\), discriminative models learn \(P(Y|X)\)
D) Generative models use labeled data, discriminative models use unlabeled data
Question 7: GAN Architecture
What are the two main components of a Generative Adversarial Network (GAN)?
A) Encoder and Decoder
B) Generator and Discriminator
C) Generator and Encoder
D) Discriminator and Decoder
Question 8: Variational Autoencoders (VAEs)
In a Variational Autoencoder (VAE), what is the role of the encoder and decoder?
A) The encoder generates data, the decoder evaluates it
B) The encoder maps input data to a latent space, the decoder generates new data from the latent space
C) The encoder classifies data, the decoder reconstructs it
D) The encoder reduces data dimensionality, the decoder increases it
Question 9: Autoregressive Models
How do autoregressive models generate data?
A) By generating all data points at once
B) By generating one data point at a time, conditioned on the previous points
C) By transforming latent variables
D) By using adversarial training
Question 10: Normalizing Flows
What is a key characteristic of Normalizing Flows?
A) They use a series of non-invertible transformations
B) They provide exact likelihood estimation and efficient sampling
C) They are primarily used for classification tasks
D) They do not require training data
Questions of Chapter 2: Understanding Generative Models
Question 6: Generative vs. Discriminative Models
What is the primary difference between generative and discriminative models?
A) Generative models classify data, discriminative models generate new data
B) Generative models learn \(P(Y|X)\), discriminative models learn \(P(X, Y)\)
C) Generative models learn \(P(X, Y)\), discriminative models learn \(P(Y|X)\)
D) Generative models use labeled data, discriminative models use unlabeled data
Question 7: GAN Architecture
What are the two main components of a Generative Adversarial Network (GAN)?
A) Encoder and Decoder
B) Generator and Discriminator
C) Generator and Encoder
D) Discriminator and Decoder
Question 8: Variational Autoencoders (VAEs)
In a Variational Autoencoder (VAE), what is the role of the encoder and decoder?
A) The encoder generates data, the decoder evaluates it
B) The encoder maps input data to a latent space, the decoder generates new data from the latent space
C) The encoder classifies data, the decoder reconstructs it
D) The encoder reduces data dimensionality, the decoder increases it
Question 9: Autoregressive Models
How do autoregressive models generate data?
A) By generating all data points at once
B) By generating one data point at a time, conditioned on the previous points
C) By transforming latent variables
D) By using adversarial training
Question 10: Normalizing Flows
What is a key characteristic of Normalizing Flows?
A) They use a series of non-invertible transformations
B) They provide exact likelihood estimation and efficient sampling
C) They are primarily used for classification tasks
D) They do not require training data
Questions of Chapter 2: Understanding Generative Models
Question 6: Generative vs. Discriminative Models
What is the primary difference between generative and discriminative models?
A) Generative models classify data, discriminative models generate new data
B) Generative models learn \(P(Y|X)\), discriminative models learn \(P(X, Y)\)
C) Generative models learn \(P(X, Y)\), discriminative models learn \(P(Y|X)\)
D) Generative models use labeled data, discriminative models use unlabeled data
Question 7: GAN Architecture
What are the two main components of a Generative Adversarial Network (GAN)?
A) Encoder and Decoder
B) Generator and Discriminator
C) Generator and Encoder
D) Discriminator and Decoder
Question 8: Variational Autoencoders (VAEs)
In a Variational Autoencoder (VAE), what is the role of the encoder and decoder?
A) The encoder generates data, the decoder evaluates it
B) The encoder maps input data to a latent space, the decoder generates new data from the latent space
C) The encoder classifies data, the decoder reconstructs it
D) The encoder reduces data dimensionality, the decoder increases it
Question 9: Autoregressive Models
How do autoregressive models generate data?
A) By generating all data points at once
B) By generating one data point at a time, conditioned on the previous points
C) By transforming latent variables
D) By using adversarial training
Question 10: Normalizing Flows
What is a key characteristic of Normalizing Flows?
A) They use a series of non-invertible transformations
B) They provide exact likelihood estimation and efficient sampling
C) They are primarily used for classification tasks
D) They do not require training data