Hello everyone,
I was wondering if anyone can help me on this. I generated an MLP neural network in MATLAB, and the output is a blackbox named net(x). I want to use this function as an objective function for a multi-objective optimization algorithm such as MOPSO. Does anybody know how to link use net(x) as in input for the optimization algorithm? Is there a special code for this?
Thanks.
You can define your objective output as a function ObjFcn.m, in which you write codes about your neural network or net(x) as well as the handle of data. Then, with selections of kinds of optimization algorithms, you can use a function like bayesopt.m or gridsearch to determine the best choice in your selections, based on your objective function.
Dear Erfan,
As YuanYuan said the output of your net can be used as an input to your Multi-Objective Particle Swarm Optimization algorithm. For more theoretical details about this issue you can find it in the following paper:
Article Multicomponent image segmentation: A comparative analysis be...
and for practical or implementation example see the following page
https://www.mathworks.com/matlabcentral/fileexchange/64371-particle-swarm-optimized-neural-network
The above is just an example where PSO is training ANN.
- Badges
- Science topic