1
$\begingroup$

I tried different values of genetic algorithm operators:

  • many crossover rates from 20% to 80%
  • many crossover rates from 1% to 20%
  • varying the population size

The study of different parameter values is called quantitative parameter tuning or sensitivity analysis. What is the difference between the two terms?

$\endgroup$
1
$\begingroup$

I would generally assume that parameter tuning is the process of finding the combination of hyperparameters (e.g., population size, crossover and mutation operators and rates, etc.) that yield the best performance on your problem. When you're thinking about the way that performance varies with parameter choice, this is the "what". What is the best choice of parameters.

Sensitivity analysis is the "how" or "how much". If I change my crossover rate from 0.9 to 1.0, how significant is the change in performance of my algorithm. Is my performance more or less stable across a wide range of choices (good) or is it highly dependent on finding this one little peak in the parameter space and every other choice is much worse (bad).

$\endgroup$
1
  • 1
    $\begingroup$ so by sensitive analysis we can have a good parameter tuning ? $\endgroup$ – fathese Nov 13 '20 at 14:28

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.