Semi-supervised Learning

Lesson goals

• Labeling cost
• Using unlabeled data
• Error propagation risks

Core idea

Combining a small labeled set with a large unlabeled set.

1. Labeling cost: understand it in context, identify inputs and outputs, and define how the result will be verified.

2. Using unlabeled data: understand it in context, identify inputs and outputs, and define how the result will be verified.

3. Error propagation risks: understand it in context, identify inputs and outputs, and define how the result will be verified.

Practical example

Consider a real application of “Semi-supervised Learning.” Start small, identify available data, and compare the system with a clear baseline before expanding it.

Check your understanding

1. Explain “Semi-supervised Learning” in your own words.
2. Give one appropriate and one misleading application.
3. What risk or limitation should be monitored?