Exam will be CLOSED BOOK, CLOSED NOTES, with NO ELECTRONIC COMPUTING DEVICES

For the exam, you should know...

 How to step through one iteration of the following algorithms:
 - Distribution of RV in BN using Variable Elimination (e.g. burglar alarm)
 - Approximate distribution of RVs in BN using MCMC (e.g. wet lawn)
 - Value of hidden RV in HMM using Filtering
 - Most Likely Sequence through HMM using Viterbi (e.g. crooked dice)
 - Approximate value of hidden RV in HMM using Particle Filter
 - Best policy through MDP using Value Iteration (e.g. cave agent)

 Short answer:
 - What are the probability axioms
 - How to obtain a conditional distribution P(A|B) from a full joint P(A,B,C,D)
 - What are the independence assumptions in a given Bayes Net
 - How to decompose a joint distribution using the chain rule
 - How to obtain P(A|e_1,e_2) from a given Bayes Net
 - What is Bayes Rule
 - What do you get when you multiply a Gaussian by a Linear Gaussian, and why
 - Compare the complexity of variable elimination with enumeration
 - What is the Markov Blanket of a node in a BN, what does this mean, and why
 - What are the axioms of utility theory
 - What is a 'monotonic' utility function
 - When does state A strictly dominate B
 - When does state A stochastically dominate B
 - How to query a Decision Network (oil drilling)