Table 2: Comparative table of adaptive self-adaptive strategies.

Technique	Algorithm	Operator	Features
1/5 Rule	GA	Mutation	-Successful mutations must be p=1/5 in a particular generation
Fixed M [41]	GA	Mutation	-Based on 1/5 Rule
Modification Fixed M [2]	GA	Mutation	-Based on Fixed M
Canonical version [3]	GA	Mutation	-Using markov chain model to converge
Self-Adaptation features [4]	GA	Mutation	Exploration phase is more beneficial high mutation Exploitation phase is more beneficial low mutation
Shaky ladder Hyperplane [5, 42, 43] Defined Functions	GA	Mutation	Reduce value to zero of mutation parameter Bolean allele represent one or two possible mutation rates Produce high values of mutation at different generations
Based on Shaky ladder [6]	GA	Mutation	Equal probability to allow mutation on different generations
Evolutionary mutation control parameter [19]	EA	Mutation	Real code evolutionary algorithm with truncation selection and with self-adaptive
Kmut-N and Kmut-P [22]	EA	Mutation	-Kmut-N mutation operator sets the values of the gens ignoring their previous values -Kmut-p mutation is a gaussian perturbation
Simulated Binary Reproduction (SBX) [7]	GA	Crossover	Crossover operator with self adaptive control parameters
SBX [8]	GA	Crossover	Produce offspring between parents using euclidian distance
Multiple Crosses per Couple [9]	GA	Crossover	Number of crossovers allowed per individual is encoded in the chromosome
Uniformly unimodal distribution crossover [10]	GA	Crossover	Rules such as preservation of statistical moments of the population distribution and degree of diversity in future offsprings
HA-ACAGA [25]	GA	Population	-Use a function to control individuals -Extra amount of pheromones are deposited on path found
Memetic Algorithm parameter control [18]	MA	Population	Define a parameter to control diversity in the population for Parallel memetic algorithm
Adaptation in population [20]	EA	Selection	-Regarding survival and reproduction are taken by the individuals themselves independently, without any central control. -Adding an adaptation mechanism allowing individuals to regulate their own selection pressure -Algorithm enables individuals to maintain estimates on the size and the fitness of the population
Immune genetic algorithm [44]	GA	Algorithm	-Describe solutions using symbolic coding and full binary tree in the chromosomes -Cross point can be selected from the intermediate nodes and the root nodes -Mutation operator is modified
Adaptive paramters for EAs [23]	EA	Algorithm	-is based on micropopulation evolutionary algorithm -main mechanims: elitism and adaptive behaviour -mechanism mixed on mutation, crossover and replacement operators -3 adaptive paramters: ambient pressure (related with population), step size por mutation operator (related with the number of variables) and crossover balance
Estimation of Distribution Algorithm [24]	EA	Learning Rate	-Adaptive learning rate with different learning rule -Chaos perturbation and elitistm strategy
Different genomes encoding by individuals [21]	EA	Imitation probability	-Each individual carries three genomes: one containing the solution to the NK landscape, one encoding imitation probability and, finally, one encoding teaching rounds -All three genomes are allowed to undergo the processes of crossover and mutation.