Lecture 7 Hyper-Heuristics I Flashcards

1
Q

What is a Hyper-Heuristic?

A

A hyper-heuristic is a search method or learning mechanism for selecting or generating heuristics to solve computationally difficult problems
Also a class of methodologies for cross-domain search

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

What are some characteristics of hyper-heuristics?

A

Operates on a search space of heuristics (neighbourhood operators) rather than directly on a search space of solutions
Aim is to take advantage of strengths and avoid weaknesses of each heuristic (operator)
No problem specific knowledge is required during the search over the heuristics (operator) space
Easy to implement, practical to deploy (easy, cheap, fast)
Existing (or computer-generated) heuristics (operators) can be used within hyper-heuristics

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

What does the framework look like for a single point search hyper-heuristic?

A

Generate initial candidate solution p
while (termination criteria not satisfied){
Select a heuristic (or subset of heuristics) h from the set of heuristics
Generate a new solution (or solutions) s by applying h to p
Decide whether to accept s or not
If (s is accepted) then p=s
}
return p;

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

What is the HyFlex framework?

A

Defines behaviour of components and arranges the interaction between them
Separation between the problem-specific and general-purpose parts, both of which are reusable and interchangeable through the HyFlex interface.

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

What does IoM stand for, and what does it do?

A

Intensity of Mutation, and it represents how many times a mutational heuristic will repeat before full termination within a hyper-heuristic. (typically goes up in intervals of 0.2, where 0 is no repeats, and 1.0 is 5 times)

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

What does DoS mean, and what does it do?

A

Depth of Search, and it represents how many times a local search heuristic will repeat before full termination within a hyper-heuristic. (typically goes up in intervals of 0.2, where 0.2 means repeating 10 times, but 1.0 repeats 20 times)

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

What does Greedy Heuristic Selection mean?

A

Apply each low level heuristic to the candidate solution and choose the one that generates the best objective value

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

What does Reinforcement Learning Heuristic Selection mean?

A

It maintains a score for each heuristic. If an improving move, then increase the score by e.g. +1, otherwise decrease the score of the heuristic e.g. -1
Concerned with how an agent ought to take actions in an environment to maximise some notion of long-term reward.

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

What does Choice Function Heuristic Selection mean?

A

The choice function maintains a record of the performance of each heuristic. Three criteria are maintained:
Its individual performance
How well it has performed with other heuristics
The elapsed time since the heuristic has been called

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

What are some misconceptions about hyper-heuristics?

A

Do not require parameter tuning
All of them are tested under a fair setting (HyFlex in this case)
Applying a hyper-heuristic to a new domain is ‘easy’
Domain specific information should not be passed to the hyper-heuristics (objective value is not a domain specific information, all others are)

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

What is the general framework of an Iterated Multi-stage Selection Hyper-heuristic?

A

Crossover operators are ignored
The parameter value of a low level heuristic is randomly reset if there is no improvement after its application, otherwise the same value is maintained.

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