Lecture 2A: Convolutional Neural Networks (Full Stack Deep Learning - Spring 2021)

In this video, we first review convolution operation, the most basic property of Convolutional Neural Networks. Then, we look at other important operations for ConvNets. Finally, we transition to looking at a classic ConvNet architecture called LeNet. 00:00 - Introduction 01:08 - Convolutional Filters 07:10 - Filter Stacks and ConvNets 11:25 - Strides and Padding 14:35 - Filter Math 21:44 - Convolution Implementation Notes 24:04 - Increasing the Receptive Field with Dilated Convolutions 27:30 - Decreasing