2003, Data mining, Knowledge Discovery and Optimization

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 $t$.

KDD

KDD Testing Machine Learning Modeling

Data Leaders Brief

Using Empirical Bayes to approximate posteriors for large "black box" estimators

Webinars

Stay Connected