Consider a random experiment E with sample space S and the probability measure P.

Assume that all events are measurable.

Let X1, X2, X3,...., Xn are random variables over S.

Now machine learning is trying to approximate f.

My doubt is what is f here? Is it P? Is it E? If not both, then what is f in this context?

  • 1
    $\begingroup$ It’s the transformation between predictors and the response variable(s). I don’t see your problem set up in a regression format, however. You just have a predictor, X. $\endgroup$ – Dave Jul 14 '19 at 12:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.