Select Page

Study Guide, Resources, and Notes for Microsoft Azure AI Fundamentals (Exam AI-900)

Study Guide, Resources, and Notes for Microsoft Azure AI Fundamentals (Exam AI-900)

Hello! Today I am collecting study materials and links to prepare for taking the AI-900 exam and I thought I’d share. I’ve added resources and notes for each exam topic. Please let me know if I can add anything.

Describe Artificial Intelligence workloads and considerations (15–20%)

Identify features of common AI workloads

Identify guiding principles for responsible AI

  • Describe considerations for fairness in an AI solution
  • Describe considerations for reliability and safety in an AI solution
  • Describe considerations for privacy and security in an AI solution
  • Describe considerations for inclusiveness in an AI solution
  • Describe considerations for transparency in an AI solution
  • Describe considerations for accountability in an AI solution
  • Resources

My Notes

What is Artificial Intelligence? In a sense, it is software that imitates human abilities like interpreting visual input, recognizing normal/abnormal events, understanding language, engaging in conversations, and making decisions based on past experiences.

  • Common AI workloads
    • Machine Learning – predictive models based on data and statistics
    • Conversational AI – AI bots that engage in human dialogue
    • Computer Vision – AI that can interpret visual input
    • Natural Language Processing – applications that can interpret written or spoken language(s)
    • Anomaly Detection – AI that can detect unusual events or events often used to take preventative actions
  • AI in Microsoft Azure
    • Platform – Data storage, compute, and services
    • Azure Machine Learning – the platform for deploying, training, and managing Machine Learning (ML) models
    • Azure Bot Services – cloud platform for developing bots
    • Cognitive Services – a suite of services that developers can use to build AI solutions
  • Risks & Challenges with AI
    • Liability for AI decisions
    • Trust in a complex system
    • AI solutions are not a fit for every problem
    • Errors have the potential to cause harm
    • Without proper data governance, data could be accidentally exposed
    • AI Bias and hallucinations can affect the results
  • Principles of Responsible AI

Describe fundamental principles of machine learning on Azure (20–25%)

Identify common machine-learning techniques

  • Identify regression machine learning scenarios
  • Identify classification machine learning scenarios
  • Identify clustering machine learning scenarios
  • Identify features of deep learning techniques

Describe core machine learning concepts

  • Identify features and labels in a dataset for machine learning
  • Describe how training and validation datasets are used in machine learning

Describe Azure Machine Learning capabilities

  • Describe the capabilities of Automated machine learning
  • Describe data and compute services for data science and machine learning
  • Describe model management and deployment capabilities in Azure Machine Learning

My Notes

Machine Learning is creating predictive models by finding relationships in data

  • Machine Learning on Microsoft Azure
  • Supervised Learning – algorithms make predictions based upon a set of examples you provide. This technique is used when you know what the outcome should look like
  • Unsupervised Learning – algorithms label the data points for you by organizing the data and/or describing its structure
  • Machine Learning Algorithms | Microsoft Azure
  • Automated Machine Learning and using ML Designer
    • Removes barriers for easier implementation – supply the data and desired model type and Azure Machine Learning will find the best model
    • Model explanations to avoid bias
    • Split data into Training and Evaluation
    • Real-time inferencing must be deployed to Kubernetes

Describe features of computer vision workloads on Azure (15–20%)

Identify common types of computer vision solution:

Identify Azure tools and services for computer vision tasks

  • Describe the capabilities of the Azure AI Vision service
  • Describe the capabilities of the Azure AI Face detection service
  • Describe the capabilities of the Azure AI Video Indexer service

My Notes

  • Applications of Computer Vision
    • Image Classification
    • Object Detection
    • Semantic Segmentation
    • Image Analysis
    • Face detection and recognition
    • Optical Character Recognition (OCR)
  • Azure Cognitive Services
    • AI application resources
      • Standalone services for specific services
      • Cognitive Services resource for multiple services
    • These are consumed by
      • REST endpoints (e.g. https:// address)
      • Authentication key
  • Image analysis with the Computer Vision service
    • Pre-trained model
    • Object detection for over 10,000 cases predefined
    • Image description and tag generation
    • Face detection and analysis
    • Content moderation
    • Text and OCR

Describe features of Natural Language Processing (NLP) workloads on Azure (15–20%)

Identify features of common NLP Workload Scenarios

  • Identify features and uses for keyphrase extraction
  • Identify features and uses for entity recognition
  • Identify features and uses for sentiment analysis
  • Identify features and uses for language modeling
  • Identify features and uses for speech recognition and synthesis
  • Identify features and uses for translation

Identify Azure tools and services for NLP workloads

  • Describe the capabilities of the Azure AI-Language service
  • Describe the capabilities of the Azure AI Speech service
  • Describe the capabilities of the Azure AI Translator service

My Notes

Natural Language Processing is test analysis, entity recognition, sentiment analysis, speech recognition, speech synthesis, Machine translation, and Semantic language modeling.

Conversational AI is a solution that enables dialog between an AI agent (aka bots) and a human. Bots can engage over web chat, email, social media, voice, and more. Bots are often engaged to assist Customer Service agents to help reduce workload.

  • Azure – Natural Language Processing
    • Text Analytics
      • Language detection
      • Key phrase extraction
      • Entity detection
      • Sentiment analysis
    • Speech
      • Text to speech
      • Speech to text
      • Speech translation
    • Translator Text
      • Text translation
    • Language Understanding
      • Custom language modeling
  • Responsible AI for Bots
    • Be transparent about what the bot can (and can’t) do
    • Make it clear that the user is communicating with a bot
    • Enable the bot to seamlessly hand-off to a human if necessary
    • Ensure the bot respects cultural norms
    • Ensure the bot is reliable
    • Respect user privacy
    • Handle data securely
    • Ensure the bot meets accessibility standards
    • Assume accountability for the bot’s actions
    • Responsible bots: 10 guidelines for developers of conversational AI – Microsoft Research
  • The QnA Maker Service
    • Define a knowledge base of question and answer pairs:
      • By entering questions and answers
      • From an existing FAQ document
      • By using built-in chit-chat
    • Consume the knowledge base from the client
      apps, including bots
  • Azure Bot Service
    • Cloud-based platform for developing and managing bots
    • Integration with LUIS, QnA Maker, and others
    • Connectivity through multiple channels

Describe features of generative AI workloads on Azure (15–20%)

Identify features of generative AI solutions

  • Identify features of generative AI models
  • Identify common scenarios for generative AI
  • Identify responsible AI considerations for generative AI

Identify capabilities of Azure OpenAI Service

  • Describe the natural language generation capabilities of Azure OpenAI Service
  • Describe code generation capabilities of Azure OpenAI Service
  • Describe image generation capabilities of Azure OpenAI Service

Sources

About The Author

Eric Rupp

I am a Data Security Technical Specialist with a concentration on Microsoft Purview, M365 Copilot, and AI. All writing and opinions are my own and do not represent any organization outside of Cloudtoso LLC.

Share This