MegaMaester

Artificial Intelligence · Module 2

Machine Learning

Module 1 explained what artificial intelligence is and where you already meet it. This module opens the box on the technique behind most modern AI: machine learning — systems that learn patterns from data instead of following rules a person wrote by hand.

You will see how machines actually learn, the main kinds of learning, why a model that aces its practice data can still fail in the real world, how the common algorithms work in plain language, how to tell whether a model is any good, and where machine learning quietly goes wrong.

Lessons
7 lessons
Estimated time
~6-8 hours
Assessment
Module quiz included

Not started yet

Lessons

Lesson 1 · 16 min

What Is Machine Learning?

Lesson 2 · 16 min

How Machines Learn from Data

Lesson 3 · 16 min

Types of Machine Learning

Lesson 4 · 16 min

Training, Testing, and Overfitting

Lesson 5 · 16 min

Common Machine Learning Algorithms

Lesson 6 · 16 min

Evaluating a Machine Learning Model

Lesson 7 · 17 min

Bias, Data, and the Limits of Machine Learning

Check your understanding

Finish the lessons, then test yourself. Your best score is saved on this device. You can also open this quiz on its own page.

Machine Learning — Module Assessment

14 questions · pass mark 75%

  1. 1.What most clearly distinguishes machine learning from traditional rule-based programming?
  2. 2.Which statement correctly describes how AI, machine learning, and deep learning relate?
  3. 3.In machine learning, what is the difference between a feature and a label?
  4. 4.What is the main lesson the Netflix Prize is often cited for?
  5. 5.Which feature of the data determines whether you can use supervised learning?
  6. 6.A model that predicts a house's sale price from its features is performing which task?
  7. 7.What does generalization refer to in machine learning?
  8. 8.A model scores 99% on training data but only 68% on the test set. What is happening?
  9. 9.Which of these algorithms is generally the most transparent and easiest to explain to a non-expert?
  10. 10.K-means clustering is an example of which kind of learning?
  11. 11.On a dataset where 1% of people are sick, a model that predicts "healthy" for everyone will have an accuracy of about:
  12. 12.In a confusion matrix, a healthy person wrongly flagged as sick is a:
  13. 13.An HR team removes the "gender" field from a resume-screening model's inputs, yet the model still favors male candidates. What is the most likely explanation?
  14. 14.A fraud-detection model performed well for two years, then its accuracy quietly dropped after fraud tactics changed — with no code change and no error message. This is a textbook case of:
Answer every question to submit.