Architecture Fundamentals Flashcards

What is the primary purpose of padding in a convolutional layer?

Answer: To preserve spatial dimensions of the input

Padding adds border pixels (usually zeros) around the input so the output feature map retains the same spatial size as the input.

In a CNN, what does 'stride' control?

Answer: How many pixels the filter moves between applications

Stride determines the step size of the filter as it slides across the input, with larger strides producing smaller output feature maps.

Which layer type in a CNN is responsible for introducing non-linearity?

Answer: Activation layer

Activation layers (e.g., applying ReLU) introduce non-linearity after convolution, enabling the network to learn complex patterns.

What distinguishes a depthwise separable convolution from a standard convolution?

Answer: It applies one filter per input channel, then combines with 1×1 convolutions

Depthwise separable convolution factorizes a standard convolution into a depthwise step (per-channel) and a pointwise 1×1 step, reducing computation.

In AlexNet, what was the purpose of Local Response Normalization (LRN)?

Answer: To normalize activations across neighboring feature maps for lateral inhibition

LRN in AlexNet suppresses activations that are weak relative to neighbors across channels, mimicking lateral inhibition seen in biological neurons.

How does a 1×1 convolution (pointwise convolution) affect a feature map?

Answer: It changes the number of channels without altering spatial dimensions

A 1×1 convolution performs a linear combination across channels at each spatial position, enabling channel dimensionality reduction or expansion.

What is the receptive field of a neuron in a CNN?

Answer: The region of the input image that influences that neuron's activation

The receptive field is the spatial extent of the original input that a given neuron 'sees', which grows with depth due to successive convolutions.