Parameter Optimization using Simulated Annealing

I am new to optimization and trying to understand the basics, so sorry if it is a dumb question. Is it possible to tune parameters (which is a search problem) of a classifier using simulated annealing or other optimization technique, just for an example optimum value of "k" in KNN (I know there is an automatic hyperparameter optimization for KNN)?

Answer 1

[Alan Weiss]

Sure, you can do anything you want. It might not be a good idea, but feel free.

Write an objective function that is, say, the cross-validation error rate for a particular parameter. If you have k as your parameter, and a cross-valudation partition c, then you might have

fun = @(k)kfoldLoss(fitcknn(X,y,'CVPartition',c,...
  'NumNeighbors',k));

The only problem with this is that k is an integer variable, and most optimizers (including simulannealbnd) work only with continuous parameters. But you could use mixed-integer ga to optimize this.

Good luck,

Alan Weiss

MATLAB mathematical toolbox documentation

[MB]

Thank you very much greatly appreciated.