The pooling layer in the convolutional neural network can reduce the size of the lower layer input. Common pooling is:

Question

Options

• AMinimum strata
• BProduct pooling layer
• CMaximum pooling layer
• DAverage pooling layer

• A
  17% (3)
• B
  6% (1)
• C
  78% (14)

Explanation

Maximum pooling (C) and average pooling (D) are the two standard pooling operations used in CNNs to downsample feature maps by selecting the maximum value or computing the mean value within each pooling window, respectively - both are well-established, widely implemented techniques that reduce spatial dimensions while retaining meaningful features.

Minimum pooling (A) is not a standard CNN operation; selecting the smallest activation typically discards the most important features (strong activations) and is counterproductive for learning. Product pooling (B) is not a recognized standard pooling method - multiplying activations within a window causes numerical instability and is not used in practice.

Memory tip: Think "M-A-X and A-V-G are the CNN standard pair" - Max keeps the strongest signal, Average keeps the balanced signal; anything else (min, product) is not part of the standard CNN toolkit.

No community discussion yet for this question.