Reasoning with Uncertainty Flashcards

1
Q

What properties must a belief measure satisfy?

A

Unit measure s must be a sample in the space
Additivity: suppose S1 and S2 are disjoint subsets of the space. Then our belief that S E S1 U S2 is the sum of our belief that s E S1 and out belief that s E s2.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What is the probability of a proposition?

A

The strength of belief that the proposition will hold.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

How to calculate conditional probability?

A

P(h | e) = P(h ^ e)/P(e)
Also posterior probabilty

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What is the chain rule?

A

P(a ^ b) = P(a)P(b |a)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What is the law of total probability?

A

P(a) = P(a | b)P(b) + P(a | -b)P(-b)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What is a bayesian network inference problem?

A

Given a bayesian network, compute the probability p(x | ly) and P(-X | ly)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

How to normalise a CPT?

A

g(X’) = G(X) / G(x) +g(-X)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What is the chain rule?

A

See slides for answer

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is local Markov property?

A

When all neighbours are independent of any other vertices

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What is the naive method of computing probabilities?

A
  1. Restrict final variable out
  2. Multiple previous table by probability table for next elvel up
  3. Sum out other variable to get final variable
  4. normalise above table to get probabilities
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What time complexity does Naive method run on?

A

O(r^n), where r is the number of rows in CPT.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What is the algorithm for variable elimination?

A

See slides

How well did you know this?
1
Not at all
2
3
4
5
Perfectly