Genetic Pattern Search and Its Application to Brain Image Classification

Table 2

Parameters setting.

IGA	Population = 20; elite number = 2; crossover rate = 0.8, mutation rate = 0.2, migration interval = 20; migration direction = forward; fitness scaling = rank; selection function = stochastic
IPS	Population = 2𝑁; poll = 2𝑁 positive basis; complete poll = yes; initial mesh size = 1; expansion factor = 2.0; contraction factor = 0.5.
GPS	GA: population = 20; elite number = 2; crossover rate = 0.8, mutation rate = 0.2, fitness scaling = rank; selection function = stochastic.; PS: population = 2𝑁; Poll = 2N positive basis; complete poll = Yes; initial mesh size = 1; expansion factor = 2.0; contraction factor = 0.5.
Termination	Maximum fitness evaluate = ∞; independent variables tolerance = 10^−4; fitness tolerance = 10^−4; stall iteration = 20; mesh precision = 10^−4.