Hi,

It will require the more computational runtime to tune the parameters using machine learning technique. I also think that your datasets do not include training and testing set with considering an over-fitting and under-fitting issues. Thus, it will be difficult for generalise your research outcome.

Also, what is the parameter that you like to tune it?

Thanks

Osman

Hi Shayma,

You could rely on a meta-optimization method using model-based optimization in order to tune PSO. 

For overall model evaluation, cross-validation can be used on your training set. Meaning that having an external test set (test set file) is not compulsory. In other words, in practice, you need to feed the training set you have to the algorithm of model-based optimization when performing cross-validation.

HTH.

Samer

Hello, 

Tuning is essentially selecting the best parameters for an algorithm to optimize its performance.

Now the question is which parameters of PSO would u like to tune using Machine Learning method?

Inspite of PSO being an efficient algorithm for solving global optimization problem, certain issues are: a)Low convergence speed; b) Use single learning pattern for all particles; c)Prone to drop into Local optimum etc.

One machine learning method is SVM.

Depending on what parameters to be controlled (Inertia weight, acceleration constant, Control exploration or something else), 

Regards,

Rebika Rai

Rebika Rai, 

The main question is: "How to tune the parameters of particles swarm optimization (PSO) algorithm using machine learning method?"

In your answer, you've mentioned SVM. Hence, could you tell us how SVM optimizes PSO method?

On the other hand, we know that PSO is one of the methods that can be used to optimize SVM, which is different from your claim. Kindly consult this article for more information:

http://www.mecs-press.org/ijmecs/ijmecs-v2-n2/IJMECS-V2-N2-05.pdf

Cheers, 

Samer

Hello,

What kind of machine learning methods do you want to use to tune the parameters of PSO? 

Best

Hello,

Samer Sarsam,

Thank you so much to have pointed out this.Yes its true, PSO is one of the methods that can be used to optimize SVM. Infact, similar work has been done by me wherein i have used SVM as a base classifier and thereby incorporated ACO, PSO and Flower Pollination (Swarm based methods) to optimize the base classifier SVM.

SVM does not optimizes PSO, and I have not mentioned that, however SVM can be used to control some of the parameters of PSO or ACO or FPAB. As i have mentioned in the last line 

"Depending on what parameters to be controlled (Inertia weight, acceleration constant, Control exploration or something else)"

The answer i has written earlier was in context to: For instance, SVM are capable of delivering high performance in terms of classification accuracy but their proficiency truly relies on an optimal choice of hyper-parameters. On the other hand, Control parameters selection of PSO has no theoretical guidance, most choices of PSO are based on experience. 

That is wat was intended..of course SVM is not an Optimization technique, it is just a learning model that can be used as a base classifier. Thank you for sharing the article.

Regards,

Rebika Rai

Hi Rebika Rai,

Thanks for clarifying your viewpoint.

Your answer now is correct and accurate as well. 

Cheers, 

Samer

Hello Samer Sarsam,

Thank you so much. Much appreciated..!!

Regards,

Rebika Rai

Hello,

I was following the question so as to discover may be a new technique to optimize PSO parameters using mechine learning method (like the SVM).

I expected some thing like : SVM with some specific training data is used to find the fittest parameters for PSO (the out put using SVM is the fittest parameters) I am wodering if such methods exist (innovative methods) ?

Best regards.

Hi Wassila,

Could you elaborate more on your proposed idea by adding extra details?

HTH.

Samer

Would you be looking for something like this?

Conference Paper Self-adaptive mutation in the Differential Evolution: Self- * search

Hello

Please, make me clear what you are going to do with the data set and how they are related to the parameters of PSO ?

Visit link for Matlab Code

https://youtu.be/OQ3T575sbI8

http://dg-algorithm.blogspot.com/

You can use meta optimization techniques.