es for an optimisation process, the evolutionary computation

es can be very powerful. The genetic algorithm has been used for

ome global optimisation problems, which is normally a binary

ational problem. For instance, the genetic algorithm has been

optimising a discovery process of the demographic history for

n imprints [Noskova, et al., 2020] and for researching how

rised nonsynonymous single nucleotide variation results in

ical effects [Korvigo, et al., 2018], where deep learning was also

ated. This study shows one important direction, i.e., whether two

n artificial intelligence can be well integrated for a better pattern

y task. In addition to binary optimisation problems, which can be

h using the genetic algorithm, there are many optimisation

which are not binary, such as the parameters of a neural network

n this case, the evolutionary algorithms can be well employed for

ng the best solution for a parameterised model. For instance,

g the subpopulation formation problem within a cancer population

genomics data is not a binary optimisation problem. Instead,

us parameter optimisation is then needed for gene expression data

et al., 2020]. The genetic programming algorithm is different

genetic algorithms and the evolutionary algorithms. It is used to

a system which has a specific speciation, which can be expressed

anguage-based and human-friendly system such as the reverse

tation used in Chapter 8 of this book. The rule discovered using

ic programming algorithm is very human-intelligence-alike. By

e specification using the breeding operators such as the insertion,

mutation, an optimised specification of a system can be found.

tic programming, hence, has been applied for automatic medical

sification for discovering intelligence rules for improving the

making accuracy [Kumar, et al., 2020]. To better understand a

l/medical system, the genetic programming may be very

in terms of interpretation and explanation. Recently, the

ary computation approaches have been combined into deep

for better performance [Baldominos , et al., 2018; Albadr, et al.,

eyaz, et al., 2020; Pham, et al., 2020; Prince, et al., 2020].