I'm looking for the name of the theorem that stated, approximately, that given any optimization algorithm that performs well when solving a specific problem, its average performance on any other conceivable problem is lower than the average performance of any other conceivable algorithm tackling the same problems.

So in short, that if an algorithm performs well in a domain, its average performance on all other domains will be subpar.

  • $\begingroup$ This isn't a general property of algorithms in general; it is, however, a property of overfit algorithms. $\endgroup$
    – Paul
    Apr 20, 2017 at 16:17
  • $\begingroup$ I don't think I believe it at that level of breadth. Is it possible that you had something narrower in mind? $\endgroup$
    – Ian
    Apr 20, 2017 at 16:17

1 Answer 1


Are you referring to the No Free lunch theorem? https://en.wikipedia.org/wiki/No_free_lunch_theorem

(I know this should be a comment, but I don't have 50 reputation points.)


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .