3 ms·
Presumably you would want the randomness to give choices that got to 0% another chance. It seems like a better way could be to use the previous success ratios a
by baddox 14y ago
Presumably you would want the randomness to give choices that got to 0% another chance. It seems like a better way could be to use the previous success ratios as a probability density function.