L11a Uncertainty171105

Uncertainty
Chapter 13
Uncertainty
Let action At = leave for airport t minutes before flight
Will At get me there on time?
Problems:
1. partial observability (road state, other drivers' plans, etc.)
2. noisy sensors (traffic reports)
3. uncertainty in action outcomes (flat tire, etc.)
4. immense complexity of modeling and predicting traffic
Hence a purely logical approach either

1. risks falsehood: “A25 will get me there on time”, or
2. leads to conclusions that are too weak for decision making:
“A25 will get me there on time if there's no accident on the bridge and it doesn't rain
and my tires remain intact etc etc.”
(A1440 might reasonably be said to get me there on time but I'd have to stay overnight
in the airport …)
Probability to the Rescue
• Probability
– Model agent's degree of belief, given the available evidence.
– A25 will get me there on time with probability 0.04
Probability in AI models our ignorance, not the true state of the

world.
State statement “With probability 0.7 I have a cavity” means:

I either have a cavity or not, but I don’t have all the necessary
information to know this for sure.
Probability
Probabilistic assertions summarize effects of
– laziness: failure to enumerate exceptions, qualifications, etc.
– ignorance: lack of relevant facts, initial conditions, etc.
Subjective probability:
• Probabilities relate propositions to agent's own state of knowledge

e.g., P(A25 | no reported accidents) = 0.06
• Probabilities of propositions change with new evidence:

e.g., P(A25 | no reported accidents, 5 a.m.) = 0.15
Making decisions under
uncertainty
Suppose I believe the following:
P(A25 gets me there on time | …) = 0.04
• Which action to choose?

Depends on my preferences for missing flight vs. time
spent waiting, etc.
– Utility theory is used to represent and infer preferences
– Decision theory = probability theory + utility theory
Syntax
• Basic element: random variable
• Similar to propositional logic: possible worlds defined by assignment

of values to random variables.
• Boolean random variables
e.g., Cavity (do I have a cavity?)
• Discrete random variables
e.g., Weather is one of <sunny,rainy,cloudy,snow>
• Domain values must be exhaustive and mutually exclusive
• Elementary proposition constructed by assignment of a value to a

random variable: e.g., Weather = sunny, Cavity = false (abbreviated
as cavity)
• Complex propositions formed from elementary propositions and
standard logical connectives e.g., Weather = sunny  Cavity = false
Syntax
• Atomic event: A complete specification of the state of the world about which the agent is
uncertain
E.g., if the world consists of only two Boolean variables Cavity and Toothache, then there are 4
distinct atomic events:
Cavity = false  Toothache = false

Cavity = false  Toothache = true
Cavity = true  Toothache = false
Cavity = true  Toothache = true
• Atomic events are mutually exclusive and exhaustive

Axioms of probability
• For any propositions A, B
true in all worlds e.g. P(a OR NOT(a))
– 0 ≤ P(A) ≤ 1
– P(true) = 1 and P(false) = 0 false in all worlds: P(a AND NOT(a))
– P(A  B) = P(A) + P(B) - P(A  B)
Think of P(a) as the number of worlds

in which a is true divided by the total number
of possible worlds.
Prior probability
• Prior or unconditional probabilities of propositions
e.g., P(Cavity = true) = 0.1 and P(Weather = sunny) = 0.72 correspond to
belief prior to arrival of any (new) evidence
• Probability distribution gives values for all possible assignments:

P(Weather) = <0.72,0.1,0.08,0.1> (normalized, i.e., sums to 1)
• Joint probability distribution for a set of random variables gives the

probability of every atomic event on those random variables
P(Weather,Cavity) = a 4 × 2 matrix of values:
Weather = sunny rainy cloudy snow

Cavity = true 0.144 0.02 0.016 0.02
Cavity = false 0.576 0.08 0.064 0.08
• Every question about a domain can be answered by the joint

distribution
Conditional probability
• Conditional or posterior probabilities
e.g., P(cavity | toothache) = 0.8
i.e., given that toothache is all I know
• (Notation for conditional distributions:

P(Cavity | Toothache) = 2-element vector of 2-element vectors)
• If we know more, e.g., cavity is also given, then we have

P(cavity | toothache,cavity) = 1
• New evidence may be irrelevant, allowing simplification, e.g.,

P(cavity | toothache, sunny) = P(cavity | toothache) = 0.8
• This kind of inference, sanctioned by domain knowledge, is crucial

Conditional probability
• Definition of conditional probability:
P(a | b) = P(a  b) / P(b) if P(b) > 0
• Product rule gives an alternative formulation:

P(a  b) = P(a | b) P(b) = P(b | a) P(a)
• A general version holds for whole distributions, e.g.,

P(Weather,Cavity) = P(Weather | Cavity) P(Cavity)
• (View as a set of 4 × 2 equations, not matrix mult.)
• Chain rule is derived by successive application of product rule:

P(X1, …,Xn) = P(X1,...,Xn-1) P(Xn | X1,...,Xn-1)
= P(X1,...,Xn-2) P(Xn-1 | X1,...,Xn-2) P(Xn | X1,...,Xn-1)
=…
= πi= 1^n P(Xi | X1, … ,Xi-1)
Inference by enumeration
• Start with the joint probability distribution:
• For any proposition a, sum the atomic events where it is true: P(a) =
Σω:ω╞a P(ω)
P(a)=1/7 + 1/7 + 1/7 = 3/7

• For any proposition φ, sum the atomic events where it is true: P(φ) =
Σω:ω╞φ P(ω)
• P(toothache) = 0.108 + 0.012 + 0.016 + 0.064 = 0.2

• Can also compute conditional probabilities:

P(cavity | toothache) = P(cavity  toothache)
P(toothache)
= 0.016+0.064
0.108 + 0.012 + 0.016 + 0.064
= 0.4
Normalization
• Denominator can be viewed as a normalization constant α
P(Cavity | toothache) = α x P(Cavity,toothache)

= α x [P(Cavity,toothache,catch) + P(Cavity,toothache, catch)]
= α x [<0.108,0.016> + <0.012,0.064>]
= α x <0.12,0.08> = <0.6,0.4>
General idea: compute distribution on query variable by fixing evidence

variables and summing over hidden variables
Typically, we are interested in
the posterior joint distribution of the query variables Y
given specific values e for the evidence variables E
Let the hidden variables be H = X - Y - E
Then the required summation of joint entries is done by summing out the hidden
variables:
P(Y | E = e) = αP(Y,E = e) = αΣhP(Y,E= e, H = h)
• The terms in the summation are joint entries because Y, E and H together
exhaust the set of random variables
• Obvious problems:
1. Worst-case time complexity O(dn) where d is the largest arity
2. Space complexity O(dn) to store the joint distribution
3. How to find the numbers for O(dn) entries
Independence
• A and B are independent iff
P(A|B) = P(A) or P(B|A) = P(B) or P(A, B) = P(A) P(B)
P(Toothache, Catch, Cavity, Weather)

= P(Toothache, Catch, Cavity) P(Weather)
• 32 entries reduced to 12;

• for n independent biased coins, O(2n) →O(n)
• Absolute independence powerful but rare
• Dentistry is a large field with hundreds of variables, none of which are

independent. What to do?
Conditional independence
• P(Toothache, Cavity, Catch) has 23 – 1 = 7 independent entries
• If I have a cavity, the probability that the probe catches in it doesn't

depend on whether I have a toothache:
(1) P(catch | toothache, cavity) = P(catch | cavity)
• The same independence holds if I haven't got a cavity:

(2) P(catch | toothache,cavity) = P(catch | cavity)
• Catch is conditionally independent of Toothache given Cavity:

P(Catch | Toothache,Cavity) = P(Catch | Cavity)
Conditional independence cont.
• Write out full joint distribution using chain rule:
P(Toothache, Catch, Cavity)

= P(Toothache | Catch, Cavity) P(Catch, Cavity)
= P(Toothache | Catch, Cavity) P(Catch | Cavity) P(Cavity)
= P(Toothache | Cavity) P(Catch | Cavity) P(Cavity)
I.e., 2 + 2 + 1 = 5 independent numbers
• In most cases, the use of conditional independence reduces the

size of the representation of the joint distribution from exponential in
n to linear in n.
• Conditional independence is our most basic and robust form of

knowledge about uncertain environments.
Bayes' Rule
• Product rule P(ab) = P(a | b) P(b) = P(b | a) P(a)
 Bayes' rule: P(a | b) = P(b | a) P(a) / P(b)
• or in distribution form
P(Y|X) = P(X|Y) P(Y) / P(X) = αP(X|Y) P(Y)
• Useful for assessing diagnostic probability from causal probability:
– P(Cause|Effect) = P(Effect|Cause) P(Cause) / P(Effect)
– E.g., let M be meningitis, S be stiff neck:

P(m|s) = P(s|m) P(m) / P(s) = 0.8 × 0.0001 / 0.1 = 0.0008
– Note: posterior probability of meningitis still very small!
Bayes' Rule and conditional
independence
P(Cavity | toothache  catch)
= αP(toothache  catch | Cavity) P(Cavity)
= αP(toothache | Cavity) P(catch | Cavity) P(Cavity)
• This is an example of a naïve Bayes model:

P(Cause,Effect1, … ,Effectn) = P(Cause) πiP(Effecti|Cause)
• Total number of parameters is linear in n
• A naive Bayes classifier computes: P(cause|effect1, effect2...)

The Naive Bayes Classifier
Imagine we have access to the probabilities of
1. P(disease)
2. P(symptoms|disease)=P(headache|disease)P(backache|disease)....
Then, the probability of a disease is computed using Bayes rule:
P(disease|symptoms) = constant x P(symptoms|disease) x P(disease)

Learning a Naive Bayes Classifier
What to do if we only have observations from a doctors office?
For instance:
flu1 headache, fever, muscle ache
lungcancer1  short breath, breast pain
flu2  headache, fever, cough
....
In general {(x1,y1), (x2,y2), (x3,y3),....}
symptoms (attributes) disease (label)
P(disease = y) = # people with disease y = fraction of people with disease y

total # of people in dataset
P(symptom_A=x_A|disease = y) = # people with disease y that have symptom A

total # people with disease y
Boosted a Naive Bayes Classifier
• This is a wonderfully simple and effective
classifier.
• Simply apply the boosting wrapper around the

naive Bayes classifier.
• If you count a data-case, you now need to

multiply the count with the weight of that data-
point.
Summary
• Probability is a rigorous formalism for uncertain
knowledge
• Joint probability distribution specifies probability
of every atomic event
• Queries can be answered by summing over
atomic events
• For nontrivial domains, we must find a way to
reduce the joint size
• Independence and conditional independence
provide the tools

L11a Uncertainty171105

Uploaded by

Copyright:

Available Formats

L11a Uncertainty171105

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

L11a Uncertainty171105

Uploaded by

Copyright:

Available Formats

Uncertainty

Hence a purely logical approach either

– Model agent's degree of belief, given the available evidence.

– A25 will get me there on time with probability 0.04

Probability in AI models our ignorance, not the true state of the

State statement “With probability 0.7 I have a cavity” means:

• Probabilities relate propositions to agent's own state of knowledge

• Probabilities of propositions change with new evidence:

• Which action to choose?

• Similar to propositional logic: possible worlds defined by assignment

• Elementary proposition constructed by assignment of a value to a

Cavity = false  Toothache = false

• Atomic events are mutually exclusive and exhaustive

– P(true) = 1 and P(false) = 0 false in all worlds: P(a AND NOT(a))

– P(A  B) = P(A) + P(B) - P(A  B)

Think of P(a) as the number of worlds

• Probability distribution gives values for all possible assignments:

• Joint probability distribution for a set of random variables gives the

Weather = sunny rainy cloudy snow

• Every question about a domain can be answered by the joint

• (Notation for conditional distributions:

• If we know more, e.g., cavity is also given, then we have

• New evidence may be irrelevant, allowing simplification, e.g.,

• This kind of inference, sanctioned by domain knowledge, is crucial

• Product rule gives an alternative formulation:

• A general version holds for whole distributions, e.g.,

• Chain rule is derived by successive application of product rule:

P(a)=1/7 + 1/7 + 1/7 = 3/7

• P(toothache) = 0.108 + 0.012 + 0.016 + 0.064 = 0.2

• Can also compute conditional probabilities:

• Denominator can be viewed as a normalization constant α

P(Cavity | toothache) = α x P(Cavity,toothache)

General idea: compute distribution on query variable by fixing evidence

Let the hidden variables be H = X - Y - E

P(Toothache, Catch, Cavity, Weather)

• 32 entries reduced to 12;

• Absolute independence powerful but rare

• Dentistry is a large field with hundreds of variables, none of which are

• If I have a cavity, the probability that the probe catches in it doesn't

• The same independence holds if I haven't got a cavity:

• Catch is conditionally independent of Toothache given Cavity:

P(Toothache, Catch, Cavity)

I.e., 2 + 2 + 1 = 5 independent numbers

• In most cases, the use of conditional independence reduces the

• Conditional independence is our most basic and robust form of

– E.g., let M be meningitis, S be stiff neck:

• This is an example of a naïve Bayes model:

• Total number of parameters is linear in n

• A naive Bayes classifier computes: P(cause|effect1, effect2...)

Imagine we have access to the probabilities of

Then, the probability of a disease is computed using Bayes rule:

P(disease|symptoms) = constant x P(symptoms|disease) x P(disease)

In general {(x1,y1), (x2,y2), (x3,y3),....}

symptoms (attributes) disease (label)

P(disease = y) = # people with disease y = fraction of people with disease y

P(symptom_A=x_A|disease = y) = # people with disease y that have symptom A

• Simply apply the boosting wrapper around the

• If you count a data-case, you now need to