BAT

[RamziAI]

Hello. I tried to use the BAT Optimization Algorithm and I expected to get approximately 0, with such benchmark functions as Rosebrock and Ackley, a number of iterations 66000. But the result I get is much higher. I wanted to know is it because of code, or by the algorithm itself?

[RamziAI]

Here I used 1000 iterations, just because the value doesn't decrease after certain point.
from NiaPy.task import StoppingTask, OptimizationType
from NiaPy.algorithms.basic import BatAlgorithm
for i in range(1000):
task = StoppingTask(D=30, nFES=1000, benchmark='ackley')
algorithm = BatAlgorithm(NP=30)
best = algorithm.run(task)
print(best[-1])

[firefly-cpp]

Hi @RamziAI

BA has some limits in solving such benchmarks. It is recommended that you approach with other well-known solvers, e. g. DE or jDE.

Just a remark - You are not running your algorithm using 1000 iterations, but you operate with number of function evaluations. You could use nGEN instead of nFES. However, I believe there will not be any special difference in final solution.