Evolutionary Computation Flashcards
Evolutionary Computation mimetize the principles of ___ to perform ___ tasks
natural evolution
computational
___ and ___ represent the essence of evolution
Variation and Selection
Living organisms act as a duality of their ___ and their ___
genotype
phenotype
Genotype is the underlying ___
genetic coding
Phenotype is the manner of ___ contained in the ___, ___ and ___ of the organism
response
behaviour, physiology and morphology
Evolution can be viewd as occurring as a succession of four mapping functions: 1 - \_\_\_ 2- \_\_\_ 3- \_\_\_ 4- \_\_\_
1- epigenesis
2- selection
3- genotypic survival
4- mutation
ESGM
ÉS Grand Master
Epigenesis function:
___
I X G -> P
Selection function:
___
P -> P
Genotypic Survival function:
___
P -> G
Mutation function:
___
G -> G
The simplest evolutionary algorithm can be viewed as a search procedure that:
1- generates ___
2- tests each for ___
3- and then generates ___
1- potencial solutions to a problem
2- suitability
3- new solutions
Exhaustive Search is certain to find the ___ as it consists of evaluating ___ but in most cases it will take ___
best solution
all possible solutions
too much time
Blind Random Search can be performed in a ___ by selecting n appropriately. However, it usually performs ___ on real-worl problems
reasonable time
poorly
Both Exhaustive and Blind Random search choose the next solution without regard to what ___
has been chosen previously
Traditional Search Procedures can ___ quickly on a ___ or ___, but run the risk of stalling at a ___
converge
maximum or minimum
saddle point
As opposed to Traditional Gradient Methods, Evolutionary Algorithms>
1- Rather than executing a ___ search, they incorporate a ___
2- Instead of utilizing ___ infromation from the response surface being searched, they utilize ___
1- poin-to-point / popularion of solutions, each individual solution competing for survival
2- gradient / random variation to explore new solutions
In the Evolutionary Algorithm the relationship between parent and offspring defines ___.
If that does not exist, then the procedure is essentially like a ___ conducted ___ at a time
inheritance
blind random search
multiple samples
Suppose the value of an offspring is related to its parent as follows:
x’ = x + N(0, sigma)
1- if sigma -> infinite then the search is like a ___
2- if sigma -> 0 then there is ___ from parent to offspring and there’s no search at all because ___
3- if 0 < sigma < infinite the procedure becomes a ___ search
1- blind random search
2- complete inheritance / each offspring merely replicates its parent
3- stochastic parallel /
In the Generate and Test Fase:
1- The population is ___
2- The parents are selected depending on the previous ___
3- Parents generate ___
1- evaluated
2- score
3- offsprings
In the Evolutionary algorithms it is key to choose what are the best:
1- ___ to represent solution
2- The ___ that are applied to those ___
3- The ___ that determine which solutions in a population will influence the creation of the next generation
3-
1- data scructures
2- search operators / data scrutcures
3- selection operators
Selection can vary im effect from ___ to ___
strong to weak
Strong Selections ensures that only the very best solutions in a population will ___
serve as parents
Weak Selection offers ___ probability that weaker solutions may also ___
nonzero
serve as parents
Some of the more common search operators include ___ that apply to a single parent, and ___ that apply to two or more parents.
mutation functions
recombinations functions
