CNN CNN - Convolutional Neural Networks Data Augmentation Techniques Questions and Answers 2

Question 1

What does the ColorJitter augmentation randomly alter in a training image?

Accepted Answer

Brightness, contrast, saturation, and hue

Answer

ColorJitter randomly perturbs brightness, contrast, saturation, and hue to make models robust to varying lighting and color conditions.

Question 2

What is RandAugment?

Accepted Answer

An automated augmentation policy that uniformly samples from a set of augmentation operations

Answer

RandAugment simplifies augmentation search by randomly applying N operations from a fixed set, each at a global magnitude M, reducing the search space significantly.

Question 3

What is Test-Time Augmentation (TTA) used for in CNN inference?

Accepted Answer

Applying multiple augmentations to test images and averaging predictions to improve accuracy

Answer

TTA applies several augmented versions of each test image, collects the model's predictions for each, and averages them to produce a more robust final prediction.

Question 4

What does AutoAugment do in the context of CNN training?

Accepted Answer

Searches for an optimal augmentation policy using reinforcement learning

Answer

AutoAugment formulates the augmentation policy search as a reinforcement learning problem, where the controller learns which transformations and magnitudes maximize validation accuracy.

Question 5

What is the 'Random Erasing' augmentation technique?

Accepted Answer

Randomly selecting a rectangle in the image and replacing it with random pixel values

Answer

Random Erasing selects a random rectangular region in the training image and replaces it with random noise or a constant value, helping the model become robust to partial occlusion.

CNN - Convolutional Neural Networks Practice Test

CNN - Convolutional Neural Networks Practice Test

CNN CNN - Convolutional Neural Networks Data Augmentation Techniques Questions and Answers 2