2015, KDD and Optimization - Data Leaders Brief

Experiment design and modeling for long-term studies in ads

The Unofficial Google Data Science Blog

OCTOBER 7, 2015

This is essentially the same as finding a truly useful objective to optimize. Recently, we presented some basic insights from our effort to measure and predict long-term effects at KDD 2015 [1]. We use this knowledge to define objective functions to optimize our ads system with a view towards the long-term.

Modeling

Modeling Experimentation Knowledge Discovery Testing

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. References [1] Omkar Muralidharan, Amir Najmi "Second Order Calibration: A Simple Way To Get Approximate Posteriors" , Technical Report, Google, 2015. [2]

KDD

KDD Testing Machine Learning Measurement

Data Leaders Brief

Experiment design and modeling for long-term studies in ads

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

Webinars

Stay Connected