Question About BAT Algorithm Convergence

Hello, Every Body

Am studying evolutionnary algorithm convergence,

now am at the BAT algorithm .

I have tested the source code of this algorithm using the well-known 23 benchmarked functions , but in vain it seems that its convergence is very bad for all the functions.

After reading its related paper i have found that:

this algorithm is tested only on 5 functions.!!

* among these 5 functions there are only 3 functions among the 23 functions of which most metaheuristic algorithms are tested.

* It seems to me that BAT algorithm, as it is never converges, !! unless there is an error in the shared source code .

here is the download link:

https://www.mathworks.com/matlabcentral/fileexchange/68981-bat-optimization-algorithm

https://www.mathworks.com/matlabcentral/fileexchange/37582-bat-algorithm-demo

I tried the two BAT codes, the two do not converge !!!!

any advice , any correction are highly appreciated

[lin lin]

Hello, I'm also studying bat algorithm recently. I found the same problem with you. Do you know the reason

[djim djim]

yes , after i studied so many algorithms !!

I know the reason !

This is what we call

"Scientific research fraud"

the algorithm has a very bad convergence rate , or also you can say : it never converge !!,

and so many researcher pretend that PSO is special case of Bat !!! ,

while, the reality it is the complete opposite.