2. Uncertainty -Introduction
What is Uncertain information?
Doubtful Information
Most intelligent systems have some degree of uncertainty associated
with them
(i.e) including human
2
3. Uncertainty -Causes
Uncertainty may occur in Knowledge Base(KB) because of the problems
with the data
Data might be missing or unavailable
Data might be present but unreliable or confusing due to
measurement errors
The representation of data may be inaccurate
Data may just be experts best guess
Data may be based on defaults
3
4. Uncertainty –How to solve this?
To solve this uncertain information we must be concerned with three
issues
How to represent uncertain data?
How to combine two or more pieces of uncertain data?
How to draw inference(knowledge) using certain data?
● Finally we need to manage uncertain information
4
5. Uncertainty –How to solve this?
Methods for managing uncertain information
Probability
Bayesian belief network
Temporal nodes
Hidden Markov models
5
6. Uncertainty –Introduction
Example for uncertainty –Wumpus world
6
1.Agent A moves right to sensor BREEZE
B ->Y
2.If BREEZE is available in the square the
neighboring squares may/may not contain PIT
3.If u take expert guess and move right to PIT game is
over
4.Instead of moving right from BREEZE If u guess and
move up the game exists –But still it is not
correct(Bcoz Poss of PIT)
5.So the Agent Moves back and moves above which is
sensor STENCH
S -> Y
7. Uncertainty –Introduction
Example for uncertainty –Wumpus world
7
6.If Stench is present in Square,its neighbor might be
Wumpus
Now,we have PIT,WUMPUS P,W near Stench Right
7.Now Lets combine Conditions
B ->Y and S -> Y
8. B ->Y then S -> N and
S -> Y then B -> N this means there is no P,W so lets
move right
So after combining 2 conditions we get a new move
8. Uncertainty –Introduction
Real time example for uncertainty
The agent wants to drive someone to the airport to catch a flight
This involves leaving home 90 minutes before the flight departs and
driving at a reasonable speed
Even though the airport is only 15 miles away ,the agent will not be
able to conclude with certainty that “plan will get us to the airport in
time”
8
9. Uncertainty –Introduction
Real time example for uncertainty
As long as
My car doesn’t break down
Run out of petrol/gas
Don’t get into an accident
No accidents on the bridge
Plane doesn’t leave early
9
11. Handling uncertain knowledge
Example :Dental diagnosis
Consider the following rule (FOL-First order logic expression)
∀p Symptom(p, Toothache) ⇒ Disease(p, Cavity)
(∀ -for all)
Toothache -pain in a tooth or teeth
Cavity -hole in your tooth
This rule is wrong .Not all patients with toothaches have cavities
Some of them have gum disease, an abscess(swelling), or one of other
several problems
11
12. Handling uncertain knowledge
∀p Symptom(p, Toothache) ⇒ Disease(p, Cavity) ∨
Disease(p, GumDisease) ∨ Disease(p, Abscess)…
We could turn the rule into causal rule
∀p Disease (p, Cavity) ⇒ Symptom(p, Toothache)
But this rule is not right either;not all cavities cause pain
12
13. Handling uncertain knowledge
Degree of Belief
The agent’s knowledge can at best provide only a degree of belief in the
relevant sentences(Based on knowledge u r believing someone)
The main tool for dealing with degrees of belief will be probability theory,
which assigns to each sentence a numerical degree of belief between 0
to 1
Ex: The probability that the patient has a cavity is 0.8 is about the agent’s
belief ,not directly about the world
13
14. Handling uncertain knowledge
Evidence
Beliefs depends on the percepts(a general rule intended to regulate
behavior or thought) that EVIDENCE the agent has received to date
These percepts constitute the evidence on which probability assertions
are based
As the agent receives new percept’s, its probability assessments are
updated to reflect the new evidence
Before the evidence is obtained, it is termed as prior or unconditional
probability
After the evidence is obtained, it is termed as posterior or conditional
probability
14
15. Uncertainty and Rational decisions
Decision theory= probability theory + utility theory
The fundamental idea of decision theory is that an agent is
rational(correct) if and only if it chooses the action that yields the highest
expected utility, averaged over all the possible outcomes of the action
This is called principle of Maximum expected utility(MEU)
Ex:if there are 5 outcomes 1 1 1 1 1(probability theory) and in that I select 1
best(utility theory)
15