Types of DL Models By Data Relationship

Fundamental Operators in Neural Networks

Matrix Multiplication

Why Linear Layers?

Activation Functions

Connections

Design - Four Quadrants and Three Dimensions

Input and Output Everywhere

Sequential-Feature Orthogonal MLP

Twisted MLP

Layer Shared Weights Residue MLP

Data Structure in Image Processing

Dropout Functions

Normalization Functions

Gradient Vanishing Problem

Initialization

Overfitting

Partially Connected Model

Convolutional Model - Basic Concepts

Problem of Fully Connected Layer

Receptive Field and Convolution

Convolution Process

Enlarge Receptive Field via Stacking Convolution

Padding

Pooling and Downsampling

Image Classification Problem Formulation

Image Data Augmentation

Convolutional Model - Advanced Modules

Residue Connection

1×1 Convolution