What is the 'Kernel' in a CNN?

What is the 'Kernel' in a CNN?

Approach: Describe the filter that slides over the input image.

In Convolutional Neural Networks, a kernel (or filter) is a small matrix of weights that slides over the input data, performing element-wise multiplication and summation to extract features (like edges or textures).

Follow-up: What is the difference between kernel size and stride?

Key concepts: convolution, feature map, stride.