Not all of ML is optimization, in the sense of minimizing some error function via computing derivatives. Some methods, like decision trees, tackle the problem i

Not all of ML is optimization, in the sense of minimizing some error function via computing derivatives. Some methods, like decision trees, tackle the problem in a completely different manner.