CNN Pooling and Activation Functions 5

Question 1

What is the output range of the Tanh activation function?

Accepted Answer

[-1, 1]

Answer

Tanh maps all real inputs to the open interval (-1, 1), with outputs saturating near ±1 for large magnitude inputs.

Question 2

In a CNN used for multi-class image classification, which activation function is typically applied at the final output layer?

Accepted Answer

Softmax

Answer

Softmax converts raw logits into a probability distribution over all classes, with each output representing the probability that the input belongs to that class.

Question 3

How does L2 pooling (also called square pooling) compute its output?

Accepted Answer

Computes the square root of the mean of squared activations in the pooling region

Answer

L2 pooling computes sqrt(mean(x²)) over the pooling region, capturing energy rather than peak activation, and is related to the RMS of the region.

Question 4

Which statement correctly describes the GELU (Gaussian Error Linear Unit) activation function?

Accepted Answer

GELU weights inputs by their Gaussian cumulative distribution, producing a smooth non-monotonic shape near zero

Answer

GELU multiplies the input by Φ(x), the standard Gaussian CDF, creating a smooth curve that gently gates activations and is used in models like BERT and GPT.

Question 5

Why might a practitioner choose strided convolutions over max pooling for downsampling in a modern CNN?

Accepted Answer

Strided convolutions allow the network to learn the optimal downsampling function rather than using a fixed rule

Answer

Strided convolutions have learnable weights, so the network can optimize how it reduces spatial resolution, whereas max pooling applies a fixed non-learned operation.

CNN - Convolutional Neural Networks Practice Test

CNN - Convolutional Neural Networks Practice Test

CNN Pooling and Activation Functions 5