MEPX
Chapter 4 of 12All chapters
  1. 1What AI means
  2. 2How machine learning works
  3. 3Training data
  4. 4Kinds of learning
  5. 5Neural networks
  6. 6Large language models
  7. 7Limits and hallucination
  8. 8Bias and fairness
  9. 9Using AI well
  10. 10Privacy and safety
  11. 11Beyond text
  12. 12AI and work

Chapter 4 of 12

Kinds of learning

Supervised, unsupervised, reinforcement.

The three families

Supervised learning uses examples with correct answers attached. Unsupervised learning finds structure in data with no labels. Reinforcement learning learns from reward and punishment over many attempts.

  • Spam detection is supervised; customer segmentation is unsupervised.
  • Game playing systems and much of the tuning of chat models use reinforcement.

Fine tuning

A general model can be adapted with a smaller specialised dataset. That is far cheaper than training from scratch and is how most task specific systems are built today.