This is not surprising to anyone, because the no free lunch theorem assumes that all datasets are randomly generated. In reality real world problems are probabl

This is not surprising to anyone, because the no free lunch theorem assumes that all datasets are randomly generated. In reality real world problems are probably drawn from some distribution, and datasets are not totally random.