Recommendations as treatments
- 1 September 2021
- journal article
- Published by Wiley in AI Magazine
- Vol. 42 (3), 19-30
- https://doi.org/10.1609/aimag.v42i3.18141
Abstract
In recent years, a new line of research has taken an interventional view of recommender systems, where recommendations are viewed as actions that the system takes to have a desired effect. This interventional view has led to the development of counterfactual inference techniques for evaluating and optimizing recommendation policies. This article explains how these techniques enable unbiased offline evaluation and learning despite biased data, and how they can inform considerations of fairness and equity in recommender systems.Keywords
This publication has 41 references indexed in Scilit:
- Counterfactual Risk MinimizationPublished by Association for Computing Machinery (ACM) ,2015
- Trustworthy online controlled experimentsPublished by Association for Computing Machinery (ACM) ,2012
- Truncated Importance SamplingJournal of Computational and Graphical Statistics, 2008
- Optimal approximation for the submodular welfare problem in the value oracle modelPublished by Association for Computing Machinery (ACM) ,2008
- Evaluating the accuracy of implicit feedback from clicks and query reformulations in Web searchACM Transactions on Information Systems, 2007
- The Nonstochastic Multiarmed Bandit ProblemSIAM Journal on Computing, 2002
- The central role of the propensity score in observational studies for causal effectsBiometrika, 1983
- THE PROBABILITY RANKING PRINCIPLE IN IRJournal of Documentation, 1977
- A Generalization of Sampling Without Replacement from a Finite UniverseJournal of the American Statistical Association, 1952
- ON THE LIKELIHOOD THAT ONE UNKNOWN PROBABILITY EXCEEDS ANOTHER IN VIEW OF THE EVIDENCE OF TWO SAMPLESBiometrika, 1933