Solving of optimisation problem with evolutionary algorithm (LM-CMA-ES)

Leave a Comment / C++, Machine Learning, Tutorials / By
\( \)

Limited-Memory CMA-ES (LM-CMA-ES)  is a variation of CMA-ES which has been tailored for large-scale optimisation with limited memory. As a warm up here, I have used this algorithm to solve the min point of the following function:

$$ z= -( 4 \times  e^{- ( (x-4)^2 +(y-4)^2 ) }+ 2 \times e^{- ( (x-2)^2 +(y-2)^2 ) } )$$

The function has the following shape (all images are svg so you have unlimited zoom):

$$ z= -( 4 \times e^{- ( (x-4)^2 +(y-4)^2 ) }+ 2 \times e^{- ( (x-2)^2 +(y-2)^2 ) } )$$
$$ z= -( 4 \times e^{- ( (x-4)^2 +(y-4)^2 ) }+ 2 \times e^{- ( (x-2)^2 +(y-2)^2 ) } )$$

and the result was what  you exactly expect, here the green line show you the path that algorithm taken to reach the min point:

Path traversed with Limited-Memory CMA-ES (LM-CMA-ES) algorithm to reach the minimum of the function.

code is available at my GitHub.

References: 1,2.

0 0 votes
Article Rating
Subscribe
Notify of
guest

This site uses Akismet to reduce spam. Learn how your comment data is processed.

0 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
0
Would love your thoughts, please comment.x
()
x