Genetic / Evolutionary Algorithms and Local Minima / Maxima

Question

I came across several posts and articles that suggested using things like simulated annealing to avoid the local min / max problem.

I do not understand why this would be necessary if you started with a sufficiently large random population.

Is this just another check to make sure that the initial population was essentially large and random enough? Or are these methods an alternative to creating a “good” initial population?

Answer 1

- - , .

Answer 2

- , / . . /. SA . , EA , , EA .

Answer 3

, . , .

A better test would be to use a statistical method to disclose information about your data set, such as variance or standard deviation.