4 ms·
Genetic Algorithms Could Be More Significant Than Machine Learning
- mohitchawlaism 10y agoWhile you might be correct (i hope you are, love to work with GA). It is generally tougher to prefer/go-with GA in industry as, 1. Two runs of the same algorithm can not be expected to produce similar results. 2. You need to be really careful with parameters set, mutation probability, recombination probability which pose real issues after a certain point. 3. If your input size is too big, say vectors of length 100, then you might want to consider discretization before processing. However, GA prove to be useful in many cases, such problems impede their extensive usage in mainstream industry.