Abstract: This work shows how to leverage causal inference to understand the behavior of complex learning systems interacting with their environment and predict the consequences of changes to the system. Such predictions allow both humans and algorithms to select changes that improve both the short-term and long-term performance of such systems. This work is illustrated by experiments carried out on the ad placement system associated with the Bing search engine.

Link to the 2012 technical report page.

@article{bottou-jmlr-2013,
  author = {Bottou, L\'eon and Peters, Jonas and Qui{\~n}onero-Candela, Joaquin 
            and Charles, Denis X. and Chickering, D. Max and Portugaly, Elon 
            and Ray, Dipankar and Simard, Patrice and Snelson, Ed},
  title = {Counterfactual Reasoning and Learning Systems: The Example of Computational Advertising},
  journal = {Journal of Machine Learning Research},
  year = {2013},
  volume = {14},
  number = {Nov},
  pages = {3207--3260},
  url = {http://leon.bottou.org/papers/bottou-jmlr-2013},
}

The following typo has been corrected on 6/9/2014:

• The first equation page 3245 should read \( V_a \approx \frac{\partial Z_a}{\partial b_a}\Big/\frac{\partial Y_a}{\partial b_a}~~ \).