To give a brief one-line description of machine learning: It is basically a function approximation given sample and hypothesis class. But this question is already tackled by statisticians (parameter as well as non-parameter). So is it a old wine in a new bottle?

  • 3
    $\begingroup$ You may want to read Leo Breiman’s paper “Statistical Modeling: The Two Cultures”. It gets at the difference in outlooks and approaches between the ML and classic stats paradigms. $\endgroup$ Nov 29, 2021 at 12:40
  • 1
    $\begingroup$ @Dave Maybe you could help us understand what the question is? With the typos, the informal language, and structural problems with the English, I just can't make sense of it. $\endgroup$
    – whuber
    Nov 29, 2021 at 14:22
  • 1
    $\begingroup$ Perhaps you meant "function approximator" rather than "function appropriator"? Please clarify. $\endgroup$
    – Galen
    Nov 29, 2021 at 14:38
  • $\begingroup$ Is "hypothesis lass" intented to be "hypothesis and loss"? $\endgroup$
    – Galen
    Nov 29, 2021 at 14:39
  • $\begingroup$ @Galen it was probably meant to be "hypothesis class". $\endgroup$ Nov 29, 2021 at 15:17


Browse other questions tagged or ask your own question.