Textbook in PDF format

Deep learning, first introduced by Hinton et al. in 2006,1 has brought great changes to the world. Modern deep learning technique relies upon deep neuron networks (DNN), which have a very long history and aim to mimic the functionality of human brains. After more than 15 years of development, it has become a great success. In many application areas, deep learning, for the first time in human history, has surpassed even human specialists. Naturally, people are eager to apply such a great and efficient tool to various application fields. The foremost application targets vision-based tasks. Historically, before the emergence of deep learning, no machine learning algorithm could match what the human eye and brain can do. Since deep learning has similar architecture as the human brain we expect it to have sim ilar performance as humans.
Preface
Introduction
Part I: Vision Applications
Vision-Based Particulate Matter Estimation
Automatic Ship Plate Recognition Using Deep Learning Techniques
Generative Adversial Network Enhanced Bearing Roller Defect Detection and Segmentation
Application of Deep Learning in Crop Stress
Part II: Signal Applications
A Mixed Pruning Method for Signal Modulation Recognition Based on Convolutional Neural Network
Broad Learning System Based on Gramian Angular Field for Time Series Classification
Denoising of Radio Modulation Signal Based on Deep Learning
A Graph Neural Network Modulation Recognition Framework Based on Local Limited Penetrable Visibility Graph
Part III: Network Applications
Study of Autonomous System Business Types Based on Graph Neural Networks
Social Media Opinions Analysis
Ethereum’s Ponzi Scheme Detection Work Based on Graph Ideas
Research on Prediction of Molecular Biological Activity Based on Graph Convolution