Uncertainty and probability Introduction Probability basics Random

Introduction
●
●
Uncertainty: A state of having limited knowledge where it is
impossible to exactly describe the existing state, a future
outcome, or more than one possible outcome.
Ubiquitous in the world
Uncertainty and probability
Igor Farkaš
Centre for Cognitive Science
DAI FMFI Comenius University in Bratislava
●
●
●
To deal with uncertainty, agents must keep track of belief states.
Probability is the measure of the likeliness that an event will occur
Alternative to logical approach
(Russell & Norvig: Artificial Intelligence, Prentice Hall, 2003)
1
Probability basics
2
Random variables
3
4
Propositions
Using the probability
5
Prior probability
6
Conditional probability
7
8
Bayes' rule
Independence
9
Probability for continuous variables
10
Interim summary
Alternative: Gaussian density
11
12
Full Bayesian learning
Example
13
Posterior probability of hypotheses
14
Example of prediction probability
15
16
MAP approximation
ML approximation
17
18
Example: Linear Gaussian model
ML parameter learning in Bayes nets
19
20