This assignment is designed to refresh any background you have on the fundamentals of machine learning (ML) and to give you a little practice in thinking about using ML for KDD.

Refer to the course intro handout for guidelines on working with other students. Remember to type your solutions and produce them only from your personal notes (not common scratch work or notes).

To draw figures, use a paint program such as Microsoft Paint, PaintShop Pro, or Adobe Illustrator, a drawing package such as Visio, AutoCAD, or CorelDraw, or (preferably) an integrated office suite such as Microsoft Office (PowerPoint) or Sun StarOffice.

1. (25 points) The Almanac Game. Problem 16.1, Russell and Norvig (p. 495-496). Make sure you turn in your original answers for all 20 questions (3 estimates each). Indicate where your estimate fell relative to your extremal (25% and 75%) estimates. And no peeking at other sources – you aren’t being graded for accuracy!
2. (25 points) Building Decision Trees. Problem 18.3, Russell and Norvig (p. 561).
3. (25 points) Representational Power of Artificial Neural Networks. Problem 19.3, Russell and Norvig (p. 597).
4. (25 points) First-Order Inductive Learning. Problem 21.6, Russell and Norvig (p. 648).

1. Reinforcement Learning. Problem 20.11, Russell and Norvig (p. 624). Give a brief (1-2 paragraph) opinion and justification.
2. Prior (Relevance) Knowledge in Learning. Problem 21.2, Russell and Norvig (p. 647).