• PEAS
• Types of Agents
• Types of Environments
• Decision Theoretic Model

I may use n-queens, checkers, chess, 8-puzzle (sliding block), rubic's cube, backgammon, tic-tac-toe as the context in any question on the exam. If you are not familiar with these, you may want to learn the basic idea of how they work.

All algorithms as described in the book

Space complexity, Time complexity. You should be able to derive and discuss these. I prefer you not just memorize them.

Optimality and Completeness. You should be able to prove and discuss these.

You should know how all of the following algorithms work. This includes being able to simulate there execution for a small problem, and discuss their optimality, completeness, space complexity and time complexity.

Depth, Breadth, Uniform cost, Iterative Deepening, Bi-directional

Greedy-best-first, A*, IDA*, Recursive Best-First Search, SMA*

Heuristics, Admissibility, Consistency, Making Heuristics

Tree vs. Graph search, Closed list.

On-line search

Search in a continuous space

Genetic Algorithms

Simulated Annealing

Particle swarm Optimization

Ply, Minimax, Terminal test, Evaluation functions, Cut-off, Quiescence search, Horizon problem, and Complexity

Min/Max search

$\alpha \beta$ pruning

$\alpha \beta$ pruning w/ random nodes no limits, that is -$\infty$ to $\infty$

$\alpha \beta$ pruning w/ random nodes and limits

• Axioms of Prob.
• Definition of Conditional Prob.
• Notation, including: P(a) means the probability P(A=True), P(A) means a vector of probabilities corresponding to all values (all two, in the binary case) of the random variable A.
• Marginalizing out variable by summing, i.e. $p(a)=\Sigma_b P(a,b)$
• Using the joint to compute arbitrary probabilities and arbitrary conditional probabilities
• Bayes Law
• Chain rule, using it both directions, that is to (1) split a joint probabilities into conditionals and marginals, and (2) form joint probabilities from conditionals and marginals.