8
$\begingroup$

Hopefully my naivety doesn't shine through in this question: Are random forests Bayesian?

Improve this question
$\endgroup$
1
  • $\begingroup$ I had the same question. $\endgroup$
    – Simon Woodward
    Apr 7, 2020 at 20:53

1 Answer 1

Reset to default
12
$\begingroup$

Not really.

Random Forests are closer to bootstrapping (there is a bootstrapping piece in the random forest fit).

Bootstrapping can be seen as a Bayes analysis with a very specific prior and only looking at summaries of the posterior, but using a specific prior (that matches frequentist output) and not looking at the whole posterior does not really follow the philosophy of Bayesian analysis.

Improve this answer
$\endgroup$

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.