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.
@article{arxiv.1209.2355,
title = {Counterfactual Reasoning and Learning Systems},
author = {Léon Bottou and Jonas Peters and Joaquin Quiñonero-Candela and Denis X. Charles and D. Max Chickering and Elon Portugaly and Dipankar Ray and Patrice Simard and Ed Snelson},
journal= {arXiv preprint arXiv:1209.2355},
year = {2013}
}