Generative Adversarial Networks Quiz Form
Test your knowledge of Generative Adversarial Networks with this 10-question quiz. Please answer all questions to the best of your ability.
Which of the following best describes the two main components of a Generative Adversarial Network (GAN)?
*
Generator and Discriminator
Encoder and Decoder
Classifier and Regressor
Optimizer and Loss
Select all typical applications of GANs.
*
Image generation
Data augmentation
Speech synthesis
Sorting algorithms
Supervised classification
In GAN training, what is the primary objective of the generator?
*
To produce samples that fool the discriminator
To classify real vs fake samples
To minimize reconstruction loss
To optimize hyperparameters
Which loss function is commonly used in the original GAN framework?
*
Binary Cross-Entropy
Mean Squared Error
Categorical Cross-Entropy
Hinge Loss
Match the GAN variant to its key characteristic.
*
Rows
Key Characteristic
WGAN
Uses Wasserstein distance
Applies conditional information
Improves stability using gradient penalty
Introduces cycle consistency
Conditional GAN (cGAN)
Uses Wasserstein distance
Applies conditional information
Improves stability using gradient penalty
Introduces cycle consistency
WGAN-GP
Uses Wasserstein distance
Applies conditional information
Improves stability using gradient penalty
Introduces cycle consistency
CycleGAN
Uses Wasserstein distance
Applies conditional information
Improves stability using gradient penalty
Introduces cycle consistency
Which of the following is a common challenge when training GANs?
*
Mode collapse
Overfitting
Vanishing gradients
All of the above
Rate your confidence in your answers to this quiz.
*
Not confident
1
2
3
4
Very confident
5
1 is Not confident, 5 is Very confident
Briefly explain what 'mode collapse' means in the context of GANs.
*
Who introduced Generative Adversarial Networks in 2014?
*
Which of the following best describes the discriminator's role?
*
Distinguish real from generated samples
Generate new data samples
Tune learning rates
Augment training data
Submit Quiz
Should be Empty: