Lecture 18 - Genetic Algorithms Flashcards

1
Q

What are two sources of variation in a gene pool?

A

Mutation

Crossover

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

How does mutation modify a gene?

A

Random modification - usually detrimental

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

How does crossover modify a gene?

A

Corresponding portions of two genes are swapped

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

What is the cycle of basic genetic alrogithms

A

Evaluation - Selection - Reproduction

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

What is the typical value of the mutation rate and why is it so low?

A

10-3

So low because it typically introduces harmful changes, rather than beneficial

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

Why would a system that only adapted through mutation not be a good idea?

A

It would adapt very slowly because most mutations would not be helpful. Much faster to combine genes

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

What is uniform one-point crossover?

A

Randomly select a point somewhere along the chromosome

Exchange the portions of the two chromosomes after the selected point

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

What is the typical crossover rate? Why is this rate chosen?

A

0.5-0.8

Allows a good number to remain unchanged and also a good number to change

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

Why would a system that only adapted using crossover not be a great idea?

A

It would adapt quickly but might never find an optimal solution - no new genes would ever be introduced

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

Crossover and Mutation each have different jobs. What are they?

A

Mutation ensures the whole solution space is searched

Crossover accelerates the search process

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

In genetic programming, the population is made up of

A

computer programs

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

In genetic programming, the evolving population is made of ______ rather than strings

A

Trees

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

How does crossover work with trees?

A

Randomly select a node in both trees

Swap the subtrees

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

In tree crossover, node selection is biased. How and why?

A

Biased to select non-terminal nodes

Swapping a terminal node is basically useless

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What two mutation methods exist for trees?

A

Subtree mutation

Point mutation

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Describe subtree mutation:

A

Mutation point in tree replaced by randomly generated subtree

17
Q

Describe point mutation, as it applies to trees:

A

Change content of the chosen node