Meta-Learning Quiz

Answers and explanations appear as you take the quiz.

1. What does meta-learning, or 'learning to learn,' primarily aim to achieve?
   • Fast adaptation to new tasks from few examples
   • Memorizing one giant dataset
   • Eliminating the need for any training
   • Compressing models to zero size
2. During meta-training, how is the data usually structured?
   • As a single task with no splits
   • As many tasks, each with a support set and a query set
   • As one image repeated
   • As unlabeled noise only
3. What does MAML learn?
   • A fixed classifier that never changes
   • A dataset of negatives
   • A model initialization that can be fine-tuned to new tasks in a few steps
   • A compression codec
4. How do Prototypical Networks classify new examples?
   • By generating new images
   • By memorizing pixel values
   • By random guessing
   • By comparing them to learned class prototypes in an embedding space
5. What is the role of MAML's inner loop?
   • To adapt the model to a specific task with a few gradient steps
   • To shuffle the dataset
   • To delete parameters
   • To set the learning rate to zero
6. How does GPT-3 perform the few-shot demonstrations in the cited paper?
   • By updating all model weights after every example
   • By including examples in the prompt without gradient updates or fine-tuning
   • By storing one prototype per class
   • By running a MAML inner loop during inference
7. In Reptile, how is the meta-initialization updated after adapting on a sampled task?
   • Replace the model with a nearest-neighbor table
   • Average class labels without updating parameters
   • Move the initialization toward the task-adapted parameters
   • Train a separate model for every example
8. What computational shortcut distinguishes Reptile from the second-order MAML formulation?
   • It requires no gradient steps
   • It stores only class prototypes
   • It trains without tasks or task-specific data
   • It can use a first-order update without computing second-order derivatives