Using Empirical Bayes to approximate posteriors for large "black box" estimators
The Unofficial Google Data Science Blog
NOVEMBER 4, 2015
Posteriors are useful to understand the system, measure accuracy, and make better decisions. Methods like the Poisson bootstrap can help us measure the variability of $t$, but don’t give us posteriors either, particularly since good high-dimensional estimators aren’t unbiased. The size and importance of these systems makes this hard.
40 
Let's personalize your content