This is basic lecture on Convolutional Neural Networks. Introduces the concept of convolution of two functions and how an input is convolved with kernels. Advantages of fewer computations in a deep network than full matrix multiplication as well as sharing of parameters.
