Login to manage your account

Please enter a valid email address.
Forgot Password?
Please enter a valid password.
OR

Don't have an account yet? Sign up

What is the purpose of dropout in a neural network?

  1. A Randomly deactivating units during training to reduce overfitting
  2. B Removing training examples
  3. C Reducing the number of layers
  4. D Speeding up inference
Answer

Randomly deactivating units during training to reduce overfitting

Dropout prevents units co-adapting and acts as an approximate ensemble. It is applied during training only, not at inference.

All Machine Learning MCQs

Login to manage your account

Please enter a valid email address.
Forgot Password?
Please enter a valid password.
OR

Don't have an account yet? Sign up as