Final Exam

Learning Objectives

All Midterm 1 Objectives found here

All Midterm 2 Objectives found here

Plus...

Bayes Nets

  1. Answer any query from a joint distribution
  2. Construct joint distribution from conditional probability tables using chain rule
  3. Construct joint distribution from Bayes net and conditional probability tables
  4. Construct Bayes net given conditional independence assumptions
  5. Identify conditional independence assumptions from real-world problems
  6. Analyze the computational and space complexity inference on joint distribution
  7. Identify independence relationships in a given Bayes net
  8. Describe why Bayes nets are used in practice
  9. Decide which sampling method is the most appropriate for an arbitrary query.
  10. Be able to work through multiple iterations of particle filtering.
  11. Understand the advantages and disadvantages of using each of the sampling methods.
  12. Give a probability estimate for your query based on the samples gathered from the Bayesian Network.
  13. Identify the advantages of Gibbs Sampling over likelihood weighted sampling.
  14. Implement particle filtering for a variety of Bayesian Networks.
  15. Apply filtering to HMM queries for each time step.

Game Theory

  1. Formulate a problem as a game
  2. Describe and compare the basic concepts in game theory
    • Normal-form game, extensive-form game
    • Zero-sum game, general-sum game
    • Pure strategy, mixed strategy, support, best response, dominance
    • Dominant strategy equilibrium, Nash equilibrium, Stackelberg equilibrium
  3. Describe iterative removal algorithm
  4. Compute equilibria for bimatrix games
    • Pure strategy Nash equilibrium
    • Mixed strategy Nash equilibrium
    • Stackelberg equilibrium (only pure strategy equilibrium is required)
  5. Understand the voting model
  6. Find the winner under the following voting rules:
    • Plurality
    • Borda count
    • Plurality with runoff
    • Single Transferable Vote
  7. Describe the following concepts, axioms, and properties of voting rules
    • Pairwise election, Condorcet winner
    • Majority consistency, Condorcet consistency, Strategy-proof
    • Dictatorial, constant, onto
  8. Understand the possibility of satisfying multiple voting rule properties

Practice Exams

Please find practice exams and solutions attached below. If you have any questions, please feel free to post on Piazza or ask during Office Hours.

Practice Final 1: blank/sol


Practice Final 2: blank/sol