CS 188: Artificial Intelligence, Fall 2006

About the Exam

Midterm Solutions

The midterm will be held on Tuesday, November 21st, in class, starting promptly at 9:40am and ending at 11am.

The midterm will be closed notes, books, laptops, and people. However, you may use a one-page cheat sheet of your own design (group design ok but not recommended) along with your cheat sheet from the first midterm if desired. You may also use a basic calculator, which will not be required, but which may be helpful.

Practice Exams

Practice exams:

Note that last term's syllabus was somewhat different in order, and so the exam topics do not align perfectly.

You can also look at old exams from previous semesters.

Review Sessions and Office Hours

Possible Midterm Topics

All topics from the first midterm may be tested in synthesis with new topics

Decisions:

• The MEU principle
• Reflex agents and policies
• Markov decision processes
• Reward functions
• Bellman Equations
• Value and policy iteration
• Be able to phrase a problem as an MDP

Reinforcement Learning:

• Exploration vs exploitation
• TD value learning / Q-learning
• Linear value function approximation

Bayes Nets:

• Conditional independence / Bayes' ball
• Basics of parameter estimation (maximum likelihood, smoothing)
• Inference by enumeration
• Inference by sampling
• Be able to draw an appropriate BN for a domain
• Decision diagrams
• Value of information

Hidden Markov Models:

• Markov and hidden Markov chains
• Forward algorithm
• Viterbi algorithm
• Particle filtering