I was just having a short discussion with a leading researcher in statistical learning theory. He told me that in his perspective NOT the issues like speed or such matter but mostly the question how to validate models or better how to validate algorithms. I wanted to ask the community about different opinions what the currently important questions in statistical learning theory and machine learning are? (On a fundamental basis, i.e. not how to optimize certain algorithms or such).


closed as primarily opinion-based by Tim Jan 28 '18 at 16:35

Many good questions generate some degree of opinion based on expert experience, but answers to this question will tend to be almost entirely based on opinions, rather than facts, references, or specific expertise. If this question can be reworded to fit the rules in the help center, please edit the question.

  • 2
    $\begingroup$ Validation can be a problem of speed too. Certain models might be too slow to validate or sue. $\endgroup$ – Aksakal Feb 29 '16 at 17:09
  • $\begingroup$ Are you looking for way to validate your model or are you looking for important research questions in machine learning in general? $\endgroup$ – Esash Nov 30 '16 at 12:39
  • $\begingroup$ @Esash I am looking for problems which cannot be done efficiently on a classical computer but are important questions in daily tasks in ML and Optimisation. $\endgroup$ – LeoW. Nov 30 '16 at 16:31