Types of Machine Learning
Choosing a Learning Type
Match the problem, data, and cost to an appropriate approach.
In this lesson
Lesson goals
- Are reliable labels available?
- Prediction, discovery, or sequential action?
- Can success be measured safely?
Core idea
Match the problem, data, and cost to an appropriate approach.
1. Are reliable labels available?: understand it in context, identify inputs and outputs, and define how the result will be verified.
2. Prediction, discovery, or sequential action?: understand it in context, identify inputs and outputs, and define how the result will be verified.
3. Can success be measured safely?: understand it in context, identify inputs and outputs, and define how the result will be verified.
Practical example
Consider a real application of “Choosing a Learning Type.” Start small, identify available data, and compare the system with a clear baseline before expanding it.
Check your understanding
- Explain “Choosing a Learning Type” in your own words.
- Give one appropriate and one misleading application.
- What risk or limitation should be monitored?

