Dive into the second part of an in-depth lecture on Convolutional Neural Networks (CNNs) from the University of Central Florida's CAP5415 course. Explore the general architecture of CNNs, including biases, parameters, and kernels. Examine the groundbreaking AlexNet model and its components such as convolutional layers, max pooling layers, and conditional layers. Learn about filter sizes, feature maps, and the optimal number of layers in a CNN. Understand the concepts of max pooling and norm normalization. Investigate the algorithm architecture, focusing on how maxpooling and feature maps work together to extract low-level features. Compare and contrast correlation and convolution in the context of CNNs. This 36-minute lecture provides a comprehensive overview of CNN architecture and functionality, building upon the foundations laid in the first part of the introduction.