Machine Learning Security with Azure Best Practices for Assessing, Securing, and Monitoring Azure Machine Learning Workloads
Implement industry best practices to identify vulnerabilities and protect your data, models, environment, and applications while learning how to recover from a security breach Key Features Learn about machine learning attacks and assess your workloads for vulnerabilities Gain insights into securing...
| Otros Autores: | , |
|---|---|
| Formato: | Libro electrónico |
| Idioma: | Inglés |
| Publicado: |
Birmingham, England :
Packt Publishing Ltd
[2023]
|
| Edición: | First edition |
| Materias: | |
| Ver en Biblioteca Universitat Ramon Llull: | https://discovery.url.edu/permalink/34CSUC_URL/1im36ta/alma991009790334906719 |
Tabla de Contenidos:
- Cover
- Title Page
- Copyright and Credits
- Foreword
- Contributors
- Table of Contents
- Preface
- Part 1: Planning for Azure Machine Learning Security
- Chapter 1: Assessing the Vulnerability of Your Algorithms, Models, and AI Environments
- Technical requirements
- Azure subscription and resources
- Reviewing the Azure Machine Learning life cycle
- ML life cycle
- Azure Machine Learning
- Introducing an ML project
- Dataset
- Training the model
- Deploying the model
- Making predictions using the deployed model
- Exploring the Zero Trust model
- Introducing the Zero Trust principles
- Explaining Zero Trust defense areas
- Assessing the vulnerability of ML assets and apps
- Identity management
- Data and data sources
- Infrastructure
- Network and endpoints
- Monitoring and maintenance
- AI/ML applications
- Summary
- Further reading
- Chapter 2: Understanding the Most Common Machine Learning Attacks
- Introducing the MITRE ATLAS Matrix
- Reconnaissance
- Resource development
- Initial access
- ML model access
- Execution
- Persistence
- Defense evasion
- Discovery
- Collection
- ML attack staging
- Exfiltration
- Impact
- Understanding ML and AI attacks
- Reconnaissance techniques
- Resource development techniques
- Initial access techniques
- ML model access techniques
- Execution techniques
- Persistence techniques
- Defense evasion techniques
- Discovery techniques
- Collection techniques
- ML attack staging techniques
- Exfiltration techniques
- Impact techniques
- Exploring Azure services involved in ML attacks
- Access
- Data
- Network
- Applications
- Compute
- Azure Machine Learning
- Summary
- Further reading
- Chapter 3: Planning for Regulatory Compliance
- Exploring Responsible AI development
- Responsible AI principles.
- Getting started with Responsible AI in your organization
- Regulatory compliance in Azure Policy for Azure Machine Learning
- Azure Security Benchmark
- Federal Risk and Authorization Management Program
- New Zealand Information Security Manual (restricted)
- NIST SP 800-53 Rev. 5
- Reserve Bank of India IT Framework for Banks v2016
- Compliance auditing and reporting
- Azure portal
- Azure Resource Graph Explorer
- Compliance automation in Azure
- Azure Blueprints
- IaC
- Summary
- Part 2: Securing Your Data
- Chapter 4: Data Protection and Governance
- Working with data governance in Azure
- Identifying challenges
- Exploring benefits
- Getting started using cloud data best practices
- Exploring Azure tools and resources
- Storing and retrieving data in Azure Machine Learning
- Connecting datastores
- Adding data assets
- Encrypting and securing data
- Encryption at rest
- Encryption in transit
- Exploring backup and recovery
- Reviewing backup options for your datastores
- Recovering your workspace
- Summary
- Chapter 5: Data Privacy and Responsible AI Best Practices
- Technical requirements
- Working with Python
- Running a notebook in Azure Machine Learning
- Installing the SmartNoise SDK
- Installing Fairlearn
- Discovering and protecting sensitive data
- Identifying sensitive data
- Exploring data anonymization
- Introducing differential privacy
- Mitigating fairness
- Fairlearn
- Working with model interpretability
- Exploring the Responsible AI dashboard
- Exploring FL and secure multi-party computation
- FL with Azure Machine Learning
- Summary
- Further reading
- Part 3: Securing and Monitoring Your AI Environment
- Chapter 6: Managing and Securing Access
- Working with the PoLP
- Authenticating with Microsoft Entra ID
- Implementing RBAC
- Working with built-in roles.
- Creating a custom role for Azure Machine Learning
- Authenticating with application identities
- Creating a service principal
- Working with managed identities
- Enhancing access security
- Conditional Access
- PIM
- Azure Key Vault
- Summary
- Chapter 7: Managing and Securing Your Azure Machine Learning Workspace
- Technical requirements
- Exploring network security
- Creating a VNet
- Securing the workspace
- Securing associated resources
- Validating connectivity
- Working with Azure Machine Learning compute
- Securing compute instances
- Securing compute clusters
- Managing container registries and containers
- Securing images with Azure Container Registry
- Working with ML endpoints
- Summary
- Chapter 8: Managing and Securing the MLOps Life Cycle
- Technical requirements
- Working with MLOps in Azure Machine Learning
- Leveraging IaC
- Combining IaC with Azure Machine Learning
- Implementing CI/CD
- Working with Azure DevOps
- Exploring event-driven workflows in Azure
- Exploring Event Grid
- Working with events in Azure Machine Learning
- Discovering event handlers in Azure
- Summary
- Chapter 9: Logging, Monitoring, and Threat Detection
- Technical requirements
- Enabling logging and configuring data retention for Azure services
- Working with Azure Monitor
- Enabling diagnostic settings
- Working with alerts
- Working with Application Insights
- Visualizing the data
- Securing resources with Microsoft Defender
- Improving our security posture
- Exploring threat management with Sentinel
- Summary
- Part 4: Best Practices for Enterprise Security in Azure Machine Learning
- Chapter 10: Setting a Security Baseline for Your Azure Machine Learning Workloads
- Setting a baseline for Azure Machine Learning
- Discovering services for added security
- Exploring an example solution architecture.
- Threat modeling for Azure Machine Learning
- Exploring the STRIDE methodology
- Getting started with the Microsoft Threat Modeling Tool
- Reviewing the shared responsibility model for cloud security
- Exploring the cloud provider responsibilities
- Reviewing customers' responsibilities
- Summary
- Index
- Other Books You May Enjoy.
