Machine Learning Flavours

3 Flavours:

  1. Supervised Learning

  2. Unsupervised Learning

  3. Reinforcement Learning

Supervised Learning - “learning with a teacher”

You give the computer examples with the correct answers. It learns the connection between the input and the output.

Example: Show photos with labels “cat” or “dog”, so it learns to tell them apart.

Common uses: spam email detection, predicting house prices.

Unsupervised Learning - “learning without a teacher”

You give the computer data without answers. It tries to find hidden patterns or groupings on its own.

Example: Give it customer shopping data, and it groups customers with similar buying habits.

Common uses: Market segmentation, finding patterns in medical data.

Reinforcement Learning - “learning by trial and error”

The computer tries something, gets feedback, and learns what works best over time.

Example: A robot tries walking, falls, adjusts, and eventually learns to walk steadily.

Common uses: Game-playing AI, self-driving cars.

Summary

Supervised: learn from correct answers

Unsupervised: find patterns without answers

Reinforcement: learn by trial and error