StackedML
Practice
Labs
Questions
Models
Pricing
Sign in
Questions
/
Deep Learning
/
Regularization & Stability
/
Dropout
← Previous
Next →
240.
Dropout at Inference Time
easy
How should predictions be made at inference time when a model was trained with dropout?
A
All neurons are kept active and their outputs are used without any scaling since dropout only affects gradient flow during the training process
B
All neurons are kept active but their outputs are scaled by (1-p) to match the expected activation magnitude seen during training
C
All neurons are weighted by their empirical activation frequency observed across training steps
D
All neurons are randomly dropped with probability p just as during training to maintain consistency
Sign in to verify your answer
← Back to Questions