This book covers various, leading-edge deep learning technologies. The author discusses new applications of deep learning and gives insight into the integration of deep learning with various application domains, such as autonomous driving, augmented reality, AIOT, 5G and beyond.
Author(s): Khaled Salah Mohamed
Series: Synthesis Lectures on Engineering, Science, and Technology
Publisher: Springer
Year: 2023
Language: English
Pages: 215
City: Cham
About This Book
Contents
1 An Introduction to Deep Learning
1.1 Machine Learning Versus Data Mining Versus Statistics
1.2 Classical Supervised Machine Learning Algorithms Versus Deep Learning
1.2.1 K-Nearest Neighbor
1.2.2 Support Vector Machine
1.2.3 Decision Tree
1.2.4 Linear Regression
1.2.5 Logistic Regression
1.2.6 Naive Bayes: Probabilistic Machine Learning
1.2.7 Random Forest
1.2.8 K-Means Clustering
1.2.9 Q-Learning
1.2.10 Deep Learning
1.3 Deep Learning Models
1.3.1 Feedforward Neural Network
1.3.2 Recurrent Neural Network
1.3.3 Convolutional Neural Network
1.3.4 Spike Neural Network
1.3.5 GANs: Generative Deep Learning
1.3.6 Transfer Learning Accelerator
1.3.7 Autoencoder
1.3.8 Comparison Between Different NN Models
1.4 Deep Learning Architectures
1.4.1 AlexNet: Classification
1.4.2 VGG: Classification
1.4.3 GoogLeNet/Inception: Classification
1.4.4 ResNets: Classification
1.4.5 MobileNets: Classification
1.4.6 RetinaNet: Classification
1.5 Deep Learning Platforms and Libraries
1.5.1 TensorFlow
1.5.2 Keras
1.5.3 Pytorch
1.6 Challenges of Deep Learning
1.6.1 Overfitting and Underfitting
1.6.2 Long Computation Time
1.6.3 Vanishing Gradient
1.6.4 Hyper-Parameters Tuning: Weights and Learning Rate
1.7 Optimization of Deep Learning
1.8 Different Types of Distance Measures in Machine Learning
1.8.1 Euclidean Distance
1.8.2 Manhattan Distance
1.8.3 Hamming Distance
1.8.4 Cosine Distance
1.8.5 Minkowski Distance
1.8.6 Jaccard Distance
1.9 Classification Evaluation Metrics
1.9.1 Confusion Metric
1.9.2 Accuracy
1.9.3 True Positive Rate (TPR)
1.9.4 False Negative Rate (FNR)
1.9.5 True Negative Rate (TNR)
1.9.6 False Positive Rate (FPR)
1.9.7 Precision
1.9.8 Recall
1.9.9 F1-Score
1.10 New Trends in Machine Learning
1.10.1 Hamiltonian Neural Networks
1.10.2 Quantum Machine Learning
1.10.3 Federated Learning
1.10.4 Self-supervised Learning
1.10.5 Zero-Shot Learning and Few-Shot Learning
1.10.6 Neurosymbolic AI
1.10.7 Binarized Neural Networks
1.10.8 Text to Video Machine Learning
1.10.9 Graph Neural Networks
1.10.10 Large Language Model (LLM)
1.11 Conclusions
References
2 Deep Learning for Autonomous Driving
2.1 Introduction
2.2 Automotive Basics
2.3 AUTOSAR
2.4 Automotive Protocols
2.4.1 Can Bus
2.4.2 FlexRay
2.4.3 Local Interconnect Network (LIN)
2.4.4 Media Oriented Systems Transport (MOST)
2.5 ADAS Technology
2.6 SAE International Standard (J3016)
2.7 Color Image Processing
2.7.1 Color Models
2.7.2 Image Processing Operations
2.8 DL-Based Object Detection Algorithms
2.8.1 You Only Look Once (Yolo): Localization
2.8.2 Single Shot Detector (SSD): Localization
2.8.3 Region-Based Convolutional Neural Networks (R-CNN): Localization
2.8.4 Fast R-CNN: Localization
2.8.5 Faster R-CNN: Localization
2.8.6 Spatial Pyramid Pooling (SPP-Net): Localization
2.8.7 Mask R-CNN: Localization
2.8.8 Comparison Between Different Object Detection Algorithms
2.9 V2X Communications and Security
2.10 Hardware Platforms for Object Detection
2.10.1 FPGA
2.10.2 High Level Synthesis (C/C++ to RTL)
2.10.3 High Level Synthesis (Python to HDL)
2.10.4 MATLAB
2.10.5 Java to VHDL
2.11 Autonomous Driving Simulator
2.12 Autonomous Driving Challenges
2.13 Conclusions
References
3 Deep Learning for IoT “Artificial Intelligence of Things (AIoT)”
3.1 Introduction
3.2 Cyber-Physical Systems
3.3 AI, Big Data and IoT
3.4 Edge Intelligence
3.5 IoT Security Requirements
3.5.1 Authorization
3.5.2 Authentication
3.5.3 Integrity
3.5.4 Confidentiality
3.5.5 Resilience to Attacks
3.6 IoT Challenges
3.7 IoT Applications
3.7.1 Industrial Automation
3.7.2 Healthcare
3.7.3 Environment Monitoring
3.7.4 Smart Buildings
3.7.5 Smart Logistics
3.8 Digital Twin
3.8.1 Digital Twin Enabling Technologies
3.8.2 Digital Twin Versus Simulation
3.8.3 How to Build a Digital Twin
3.9 New Trends for IoT/AIoT
3.9.1 IIoT
3.9.2 LLM for AIoT
3.9.3 IoUT
3.9.4 IoBNT
3.9.5 IoIV
3.9.6 TinyML
3.10 Conclusions
References
4 Deep Learning for Spatial Computing: Augmented Reality and Metaverse “the Digital Universe”
4.1 Introduction
4.2 Immersive Technologies: Virtual Reality, Augmented Reality, Augmented Virtuality, Mixed Reality
4.2.1 Types of Augmented Reality
4.2.2 Augmented Reality Key Components
4.2.3 Metaverse
4.2.4 Interaction with Virtual Digital Representation
4.3 Virtual/Augmented Reality Applications
4.3.1 Augmented Reality in Digital Learning
4.3.2 Augmented Reality in Tourism
4.3.3 Augmented Reality in Games
4.3.4 Augmented Reality in eCommerce
4.3.5 Augmented Reality in Consumer Electronic
4.3.6 Augmented Reality in Manufacturing
4.3.7 Augmented Reality in Health
4.3.8 Augmented Reality in Construction
4.3.9 Augmented Reality in Fashion
4.4 Conclusions
References
5 Deep Learning for 5G and Beyond
5.1 Introduction
5.2 Deep Learning Applications in 5G/6G
5.3 Deep Learning Applications at Physical Layer
5.3.1 Channel Coding
5.3.2 Synchronization
5.3.3 Positioning
5.3.4 Channel Estimation/Prediction
5.3.5 Beamforming Design
5.3.6 Optimization
5.3.7 Spectrum Sensing, Sharing and Management
5.3.8 Interference Management
5.3.9 Error Detection and Correction
5.4 Deep Learning Applications at MAC Layer
5.4.1 Flexible Duplex
5.4.2 Power Management
5.4.3 Resource Allocation and Management
5.4.4 Modulation and Coding Scheme Selection
5.4.5 Scheduling
5.4.6 Link Evaluation
5.5 Deep Learning Applications at Network Layer
5.5.1 Routing
5.5.2 Anomaly Detection
5.5.3 Traffic Prediction
5.6 Deep Learning Applications at Application Layer
5.6.1 Performance Management
5.7 ML Deployment for 5G/6G: Case-Studies
5.7.1 BER Estimator
5.7.2 Blockchain for Security
5.8 Conclusions
References
6 Python for Deep Learning: A General Introduction
6.1 Introduction
6.1.1 What Is Python?
6.1.2 Why Should You Learn Python?
6.1.3 Python Applications: What Can You Do with It?
6.1.4 Python Programming Environment and Tools: What Do You Need to Write Programs in Python?
6.1.5 Integrated Development Environment (IDE)
6.1.6 The Big Picture of Any Program
6.1.7 Create Your First Program in Python: Hello World Program
6.1.8 Python Versus Java
6.1.9 Python Versus C++
6.2 Data Types
6.2.1 Numbers and Functions of Numbers
6.2.2 Strings and Functions of Strings
6.2.3 List and Functions of List
6.2.4 Tuples
6.2.5 Dictionary
6.2.6 Classes and Objects
6.3 Inputs
6.4 External Functions: User-Defined Functions
6.5 Control: If Statement, While Loop, For Loop
6.5.1 If/If Else/If Elif Else Statement
6.5.2 While Loop
6.5.3 For Loop
6.6 Reading and Writing from External Files
6.7 Modules and pip
6.8 Debugging
6.8.1 Comments
6.8.2 Printing Quotations
6.8.3 Assert
6.8.4 Try-Except
6.9 Data Visualization
6.10 Database
6.11 GUI
6.12 From Python to .exe
6.13 Images
6.14 Web Browser
6.15 Android
6.16 Python to HDL
6.17 Python for Machine Learning
6.18 Conclusions
References
Index
