Using Empirical Bayes to approximate posteriors for large "black box" estimators
The Unofficial Google Data Science Blog
NOVEMBER 4, 2015
Limitations Second order calibration, like ordinary calibration, is intended to be easy and useful, not comprehensive or optimal, and it shares some of ordinary calibration’s limitations. Both methods can be wrong for slices of the data while being correct on average, since they only use the covariate information through $t$.
Let's personalize your content