StackedML
Practice
Labs
Questions
Models
Pricing
Sign in
Questions
/
Deep Learning
/
Architectures (Conceptual)
/
CNN basics
← Previous
Next →
633.
Purpose of Pooling Layers
easy
What is the purpose of a pooling layer in a CNN?
A
It combines feature maps from different filters by summing their values at each spatial position
B
It reduces the spatial dimensions of feature maps, providing spatial invariance and reducing computational cost for subsequent layers
C
It applies a nonlinear activation to each spatial position, introducing the nonlinearity needed for complex feature detection
D
It normalizes the activations within each spatial region to zero mean, improving gradient flow through the network during the forward pass
Sign in to verify your answer
← Back to Questions