Beta. Content is under active construction and has not been peer-reviewed. Report errors on
GitHub
.
Disclaimer
Theorem
Path
Curriculum
Paths
Demos
Diagnostic
Search
Quiz Hub
/
Data Augmentation Theory
Data Augmentation Theory
3 questions
Difficulty 3-5
View topic
Foundation
0 / 3
1 foundation
2 intermediate
Adapts to your performance
1 / 3
foundation (3/10)
conceptual
Data augmentation creates new training examples by applying label-preserving transformations. What's the primary theoretical justification?
Hide and think first
A.
It encodes invariances (e.g., a cat is still a cat if you flip the image), effectively expanding the training distribution and acting as regularization
B.
It reduces the number of parameters needed in the model, since augmented data is more informative
C.
It randomizes training to prevent the model from memorizing specific examples, like dropout
D.
It increases the entropy of the loss function, making optimization easier
Submit Answer