csit 5220 lecture 04: building models l objective n discuss practical considerations in model...
TRANSCRIPT
CSIT 5220
Lecture 04: Building Models
Objective Discuss practical considerations in model building
Reading Jensen and Nielsen, Chapter 3
Outline Catching the structure Determining probabilities Reducing the number of parameters Other Issues
Page 1
CSIT 5220
Catching the Structure
Identify the variables
Hypothesis variables Those whose values are not directly observed, and we wish to estimate
Information variables Those whose values are observed directly observed and contain information
about the hypothesis variables
Mediating variables Those that provide information channels between the information variables and
the hypothesis variables
Build the structure
Begin with causality
Consider conditional independence
Page 2
CSIT 5220
Example: Sore ThroatPage 3
Angina is chest pain or discomfort that occurs when an area of your heart muscle doesn't get enough oxygen-rich blood.
CSIT 5220
Example: Sore Throat
Check conditional independence
Fever independent of Spots given Angina?
Page 4
CSIT 5220
Example: Infected MilkPage 5
CSIT 5220
Example: Infected MilkPage 6
CSIT 5220
Example: Infected MilkPage 7
CSIT 5220
Example: Insemination of a cowPage 8
Insemination is the process of impregnating the femalemale
CSIT 5220
Example: Insemination of a cowPage 9
CSIT 5220
Why Mediating VariablesPage 10
CSIT 5220
Example: Simplified Poker GamePage 11
CSIT 5220
Example: Simplified Poker GamePage 12
CSIT 5220Page 13
P(OH0), P(FC|OH0), P(OH1|OH0, FC), P(SC|OH1), P(OH|OH1, SC) are easier to obtain than P(OH|FC, SC)
Will see later
CSIT 5220
SummaryPage 14
CSIT 5220
Outline
Outline Catching the structure Determining probabilities Reducing the number of parameters Other Issues
Page 15
CSIT 5220Page 16
CSIT 5220Page 17
CSIT 5220
P(Test|Inf) from Sensitivity & Specificity of Test
P( Test=y | Inf=y ) = sensitivity
P( Test=y| inf=n) = 1-specificity
CSIT 5220Page 19
CSIT 5220Page 20
CSIT 5220Page 21
CSIT 5220Page 22
CSIT 5220Page 23
CSIT 5220Page 24
CSIT 5220
Stud Farm Inference Results
CSIT 5220Page 26
CSIT 5220Page 27
CSIT 5220Page 28
CSIT 5220Page 29
CSIT 5220Page 30
CSIT 5220
Outline
Outline Catching the structure Determining probabilities Reducing the number of parameters Other Issues
Page 31
CSIT 5220Page 32
CSIT 5220Page 33
CSIT 5220Page 34
CSIT 5220Page 35
CSIT 5220Page 36
CSIT 5220Page 37
CSIT 5220Page 38
CSIT 5220
Outline
Outline Catching the structure Determining probabilities Reducing the number of parameters Other issues
Page 39
CSIT 5220
Logical Constraints
Sometimes relationships among variables are undirected
Page 40
CSIT 5220
Logical ConstraintsPage 41
CSIT 5220Page 42
CSIT 5220Page 43
CSIT 5220
Probabilities need not be exact to be useful
Some people have shied away from using Bayes nets because they imagine
they will only work well, if the probabilities upon which they are based are
exact.
This is not true. It turns out very often that approximate probabilities, even
subjective ones that are guessed at, give very good results. Bayes nets are
generally quite robust to imperfect knowledge.
Often the combination of several strands of imperfect knowledge can allow us
to make surprisingly strong conclusions.
In some cases, we have no choice but trust judgments by experts
If we can trust decisions by experts, then we can trust the probability
assessments by experts
Bayes nets can help experts make better decisions, albeit subjective.
Page 44
CSIT 5220
Causal Conditional Probabilities are easier to estimate than the reverse
Studies have shown people are better at estimating probabilities "in
the forward direction".
For example, doctors are quite good at giving the probability
estimates for "if the patient has lung cancer, what are the chances
their X-ray will be abnormal?",
rather than the reverse, "if the X-ray is abnormal, what are the
chances of lung cancer being the cause?" (Jensen96)
Page 45