InfoNCE and SimCLR Objectives Quiz

Answers and explanations appear as you take the quiz.

1. What does the InfoNCE objective encourage a model to do?
   • Make a query more similar to its true positive than to many negatives
   • Memorize exact pixel values of each image
   • Minimize the number of model parameters
   • Predict the next word in a sentence
2. In SimCLR, how is a positive pair created?
   • By pairing an image with its human label
   • By applying two different random augmentations to the same image
   • By choosing two completely different images
   • By rotating the entire dataset
3. What role does the temperature τ play in NT-Xent / InfoNCE?
   • It sets the learning rate
   • It controls the batch size automatically
   • It scales the similarity scores; lower τ sharpens the distribution and emphasizes hard negatives
   • It determines the number of layers in the encoder
4. What happens to SimCLR's projection head after pretraining?
   • It becomes the final classifier
   • It is frozen and used as the encoder
   • It is duplicated for inference
   • It is discarded; the representation before the head is used for downstream tasks
5. Why did SimCLR rely on large batch sizes?
   • To provide many negative examples within each training step
   • To reduce memory usage
   • To avoid using augmentations
   • To eliminate the need for a temperature parameter
6. In SimCLR, which examples usually act as negatives for one anchor view?
   • The matching view of the same image
   • The other augmented views in the batch that come from different images
   • Only examples with the same class label
   • Views produced by a second encoder trained on labels
7. Why does SimCLR use a nonlinear projection head before computing its contrastive loss?
   • It replaces the image encoder
   • It supplies class labels during pretraining
   • The learned projection helps the contrastive objective while the representation before the head can transfer downstream
   • It removes the need for data augmentation
8. For an InfoNCE anchor, what should happen to the positive similarity relative to sampled negatives?
   • Every negative should have a higher score than the positive
   • All scores must be identical
   • The positive is excluded from the objective
   • The positive should receive a higher score than the negatives