Biased selection

[Azzaare]

Not only bad variables should be selected but occasionally some good variables too to avoid local extrema.

@richoux I mention you here as it is also related to renewal of GHOST strategy.

[richoux]

There are several ways to handle that, for instance with a flexible tabu list or some randomness in the variable selection heuristic. Any idea which direction to favor first?

[Azzaare]

I was planning to combine it with a heuristic selection such as roulette selection (as in GA).

1. Select a random non-tabu (or equivalent) variable
2. With the roulette selection threshold probability select or not that variable
3. If not selected in 2, with a low/bias probability, select it nonetheless
4. If not selected repeat from 1.

Roughly something like that.