I read very often that Bayesian algorithms work well on small datasets. Why is that? I think it is because they might generalize more, but why is that?
See also Investigating the use of Bayesian networks for small dataset problems.
$\begingroup$
$\endgroup$
-
$\begingroup$ Hi Jennifer. You're already linking us to an answer to this question (or a similar question). So, which answer are you exactly looking for? The usual explanation to your question is that Bayesian algorithms are, in some way, based on the Bayes' rule, so we can incorporate some form of knowledge of the unknown parameters by specifying e.g. the prior over the parameters. I may be able to provide a more formal answer, with examples, but I don't know when. $\endgroup$ – nbro♦ Nov 24 '19 at 0:43
