What is it?
Machine Learning is a way for computers to learn patterns from examples.
A trained model applies those patterns to new inputs to classify, predict, rank, or recommend.
01 · AI Foundations · Topic 02
What is Machine Learning?
Machine Learning lets computers learn patterns from examples instead of receiving every rule by hand.
Think of it like...
Think of it like teaching with worked examples instead of handing over a rulebook. Show many examples of spam and not spam; the learner uses those examples to classify a new message.
How it works
Training turns examples into a model. Inference uses the learned model on new data.
1 · Training dataShow examplesInputs and expected outcomes provide practice.
2 · LearningAdjust settingsThe model changes internal values to reduce errors.
3 · ModelKeep patternsTraining leaves a reusable set of learned patterns.
4 · New inputTry a new caseThe model receives data it did not train on.
5 · PredictionReturn a resultThe output may be a label, score, rank, or forecast.
Real-world examples
Everyday example · Spam filter
Input: A new email
System: Patterns learned from spam and normal email
Output: Spam or not spam
Everyday example · Recommendations
Input: A user history
System: Similar behavior helps rank possible items
Output: A short recommendation list
Business example · Demand forecast
Input: Past sales and calendar data
System: The model finds patterns in demand
Output: A forecast for future stock needs
What it is NOT
Machine Learning ≠ Artificial Intelligence
Machine Learning: One approach inside the broader AI field
Artificial Intelligence: The broader field, including learned and rule-based systems
Machine Learning ≠ Deep Learning
Machine Learning: Includes many methods such as trees and linear models
Deep Learning: A Machine Learning approach built with multi-layer neural networks
Machine Learning ≠ Fixed-rule programming
Machine Learning: Learns patterns from examples
Fixed-rule programming: Uses rules explicitly written by a developer
Related concepts
Artificial Intelligence ↓ Machine Learning ↓ Deep Learning
Machine Learning also connects to evaluation because predictions need to be tested.
Explore next
Remember this
Machine Learning learns patterns from examples and applies them to new inputs.
Independent video · Azure Neural voice