Deep Learning with Azure: Building and Deploying Artificial Intelligence Solutions on the Microsoft AI Platform

Table of Contents
About the Authors
About the Guest Authors of Chapter 7
About the Technical Reviewers
Acknowledgments
Foreword
Introduction
Part I: Getting Started with AI
	Chapter 2: Overview of Deep Learning
		Common Network Structures
			Convolutional Neural Networks
			Recurrent Neural Networks
			Generative Adversarial Networks
			Autoencoders
		Deep Learning Workflow
			Finding Relevant Data Set(s)
			Data Set Preprocessing
			Training the Model
			Validating and Tuning the Model
			Deploy the Model
			Deep Learning Frameworks & Compute
			Jump Start Deep Learning: Transfer Learning and Domain Adaptation
			Models Library
		Summary
	Chapter 3: Trends in Deep Learning
		Variations on Network Architectures
			Residual Networks and Variants
			DenseNet
			Small Models, Fewer Parameters
			Capsule Networks
			Object Detection
			Object Segmentation
			More Sophisticated Networks
			Automated Machine Learning
		Hardware
			More Specialized Hardware
			Hardware on Azure
			Quantum Computing
		Limitations of Deep Learning
			Be Wary of Hype
			Limits on Ability to Generalize
			Data Hungry Models, Especially Labels
			Reproducible Research and Underlying Theory
		Looking Ahead: What Can We Expect from Deep Learning?
			Ethics and Regulations
		Summary
	Chapter 1: Introduction to Artificial Intelligence
		Microsoft and AI
		Machine Learning
		Deep Learning
			Rise of Deep Learning
			Applications of Deep Learning
		Summary
Part II: Azure AI Platform and Experimentation Tools
	Chapter 4: Microsoft AI Platform
		Services
			Prebuilt AI: Cognitive Services
			Conversational AI: Bot Framework
			Custom AI: Azure Machine Learning Services
			Custom AI: Batch AI
		Infrastructure
			Data Science Virtual Machine
			Spark
			Container Hosting
			Data Storage
		Tools
			Azure Machine Learning Studio
			Integrated Development Environments
			Deep Learning Frameworks
		Broader Azure Platform
		Getting Started with the Deep Learning Virtual Machine
			Running the Notebook Server
		Summary
	Chapter 5: Cognitive Services and Custom Vision
		Prebuilt AI: Why and How?
		Cognitive Services
		What Types of Cognitive Services Are Available?
			Computer Vision APIs
				How to Use Optical Character Recognition–
				How to Recognize Celebrities and Landmarks
		How Do I Get Started with Cognitive Services?
		Custom Vision
			Hello World! for Custom Vision
			Exporting Custom Vision Models
		Summary
Part III: AI Networks in Practice
	Chapter 6: Convolutional Neural Networks
		The Convolution in Convolution Neural Networks
			Convolution Layer
			Pooling Layer
			Activation Functions
				Sigmoid
				Tanh
				Rectified Linear Unit
		CNN Architecture
		Training Classification CNN
		Why CNNs
		Training CNN on CIFAR10
		Training a Deep CNN on GPU
			Model 1
			Model 2
			Model 3
			Model 4
		Transfer Learning
		Summary
	Chapter 7: Recurrent Neural Networks
		RNN Architectures
		Training RNNs
		Gated RNNs
		Sequence-to-Sequence Models and Attention Mechanism
		RNN Examples
			Example 1: Sentiment Analysis
			Example 2: Image Classification
			Example 3: Time Series
		Summary
	Chapter 8: Generative Adversarial Networks
		What Are Generative Adversarial Networks?
		Cycle-Consistent Adversarial Networks
			The CycleGAN Code
			Network Architecture for the Generator and Discriminator
			Defining the CycleGAN Class
			Adversarial and Cyclic Loss
		Results
		Summary
Part IV: AI Architectures and Best Practices
	Chapter 9: Training AI Models
		Training Options
			Distributed Training
			Deep Learning Virtual Machine
			Batch Shipyard
			Batch AI
			Deep Learning Workspace
		Examples to Follow Along
			Training DNN on Batch Shipyard
				Hyperparameter Tuning
				Distributed Training
				Training CNNs on Batch AI
				Hyperparameter Tuning and Distributed Training
				Variation of Batch AI with Python SDK
			Azure Machine Learning Services
			Other Options for AI Training on Azure
		Summary
	Chapter 10: Operationalizing AI Models
		Operationalization Platforms
			DLVM
			Azure Container Instances
			Azure Web Apps
			Azure Kubernetes Services
			Azure Service Fabric
			Batch AI
			AZTK
			HDInsight and Databricks
			SQL Server
		Operationalization Overview
		Azure Machine Learning Services
		Summary
Appendix: Notes
	Chapter 1
	Chapter 2
	Chapter 3
	Chapter 4
	Chapter 5
	Chapter 6
	Chapter 7
	Chapter 8
	Chapter 9
	Chapter 10
Index