Microsoft Azure AI Fundamentals (AI-900) Flashcards¶
37 cards from this guide's fact sheet and notes. Click a question to reveal the answer.
To study with spaced repetition, download flashcards.csv and import it into Anki or Quizlet (comma separated, first row is the header). Back to the study guide.
Ai concepts¶
Types (ai concepts)
Supervised Learning: Train with labeled data (classification, regression)
Examples (ai concepts)
Fraud detection in financial transactions
Common Tasks (ai concepts)
Image Classification: Categorize entire images
Process (ai concepts)
Ingest: Collect data from various sources
Capabilities (ai concepts)
Form recognition and data extraction
Considerations (ai concepts)
Identify and mitigate algorithmic bias
Characteristics (ai concepts)
Inspired by human brain structure
Applications (ai concepts)
Image recognition and computer vision
Azure ai services¶
REST APIs (azure ai services)
Simple HTTP-based interfaces
SDKs (azure ai services)
Available for .NET, Python, Java, JavaScript, Go
Containers (azure ai services)
Deploy on-premises or edge
Custom Models (azure ai services)
Train with your own data
Multi-region (azure ai services)
Global availability
Security (azure ai services)
Azure AD integration, key-based auth
Free Tier (azure ai services)
Limited transactions for testing
Standard Tier (azure ai services)
Pay-as-you-go pricing
Commitment Tier (azure ai services)
Discounted rates for committed usage
Precision (azure ai services)
Accuracy of predictions
Recall (azure ai services)
Coverage of correct items
AP (Average Precision) (azure ai services)
Overall model quality
Intents (azure ai services)
User's goal (e.g., BookFlight)
Entities (azure ai services)
Key information (e.g., destination, date)
Utterances (azure ai services)
Example phrases
Features (azure ai services)
Contextual and action characteristics
Exploration (azure ai services)
Try different options
Exploitation (azure ai services)
Use best-known option
Apprentice Mode (azure ai services)
Shadow existing system
Sensitivity (azure ai services)
Detection threshold (0-99)
Custom Interval (azure ai services)
Data point frequency
Seasonality (azure ai services)
Periodic patterns
Responsible ai ml fundamentals¶
Completeness (responsible ai ml fundamentals)
No missing values
Consistency (responsible ai ml fundamentals)
Standardized format
Relevance (responsible ai ml fundamentals)
Useful features
Quantity (responsible ai ml fundamentals)
Sufficient examples
Demographic Parity (responsible ai ml fundamentals)
Equal positive rate across groups
Equal Opportunity (responsible ai ml fundamentals)
Equal true positive rate
Equalized Odds (responsible ai ml fundamentals)
Equal TPR and FPR across groups