Neural Network Fundamentals for Image Data

Explain why convolutional layers are more effective than fully-connected layers for image processing tasks.

Describe the role of non-linear activation functions (e.g., ReLU) in enabling neural networks to learn complex patterns.

Explain the mathematical operation of a discrete convolution and its effect on an input feature map.

Describe how pooling layers (e.g., Max Pooling) create translation invariance and reduce computational load.

Trace the flow of data through a simple Convolutional Neural Network (CNN) to produce a set of feature maps.

Explain the concept of a receptive field and how it expands in deeper layers of a CNN.