Meeting 16: Uncertainty and Probability

Reading: AIAMA 13.1-13.4

Thus far, all of our environments have been deterministic [1]. Although useful for applying AI in the virtual world, this is clearly not a good model of the real world. This reading will introduce the third big idea for the course, the representation of uncertainty and reasoning under uncertainty. Our tool of choice here is probability theory.

Questions you should be able to answer after reading are:

  1. How is probability used to represent uncertainty?
  2. What is a random variable?
  3. What is decision theory?
  4. Define a probability model.
  5. What is the joint probability?
  6. What is probabilistic inference?
  7. What is a prior probability?
  8. What is a likelihood?
  9. What is a posterior probability?
  10. What is the evidence?
  11. What is a probability density function?
  12. What is a probability mass function?
  13. How does one do inference using marginalization?
  14. What does it mean for two random variables to be independent?

[1] except for our discussion of chance nodes in games