Convolution is a mathematical operation on two functions (f and g) that creates a third function (f * g) that expresses how the shape of one is modified by the other in mathematics (specifically, functional analysis). If one of the functions participating in the fold is considered to be the indicator function of the interval, the fold can also be considered to be a 'sliding average' promotion. The idea of applying the convolutional operation to image data is not novel or specific to convolutional neural networks. A convolution is simply the application of a filter to an input that results in an activation. In computer vision, it's a common technique.
2021-09-21