Autonomous Parameter Balance in Population-Based Approaches: A Self-Adaptive Learning-Based Strategy

. 2024 Jan 31;9(2):82. doi: 10.3390/biomimetics9020082

Algorithm 1 Proposed LBLP

1:
The driver: set initial parameters required for movement operators
2:
Regression Model: set internal parameters, DSfit, DSprob, DSrank, DSsol
3:
The driver: generate initial population
4:
Roulette selector: select initial scheme to perform
5:
while $(i \leq M a x i m u m I t e r a t i o n)$ do
6:
The driver: perform intensification family of operators
7:
The driver: perform diversification family of operators
8:
Regression Model: store and check values for DSfit
9:
if best value was reached performing scheme selected by Roulette selector then
10:
Regression Model: check and update values for DSrank
11:
end if
12:
if threshold $β$ is met then
13:
Regression Model: perform regression analysis
14:
Regression Model: update DSsol
15:
Regression Model: check $M I N (D S s o l)$
16:
Regression Model: update DSprob
17:
end if
18:
if threshold $α$ is met then
19:
Roulette selector: select scheme to perform
20:
if check number of agents by the scheme selected then
21:
Roulette selector: balance the population
22:
end if
23:
end if
24:
end while