Human Interaction with Recommendation Systems

Sven Schmit, Carlos Riquelme
Proceedings of the Twenty-First International Conference on Artificial Intelligence and Statistics, PMLR 84:862-870, 2018.

Abstract

Many recommendation algorithms rely on user data to generate recommendations. However, these recommendations also affect the data obtained from future users. This work aims to understand the effects of this dynamic interaction. We propose a simple model where users with heterogeneous preferences arrive over time. Based on this model, we prove that naive estimators, i.e. those which ignore this feedback loop, are not consistent. We show that consistent estimators are efficient in the presence of myopic agents. Our results are validated using extensive simulations.

Cite this Paper

BibTeX
@InProceedings{pmlr-v84-schmit18a, title = {Human Interaction with Recommendation Systems}, author = {Schmit, Sven and Riquelme, Carlos}, booktitle = {Proceedings of the Twenty-First International Conference on Artificial Intelligence and Statistics}, pages = {862--870}, year = {2018}, editor = {Storkey, Amos and Perez-Cruz, Fernando}, volume = {84}, series = {Proceedings of Machine Learning Research}, month = {09--11 Apr}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v84/schmit18a/schmit18a.pdf}, url = {https://proceedings.mlr.press/v84/schmit18a.html}, abstract = {Many recommendation algorithms rely on user data to generate recommendations. However, these recommendations also affect the data obtained from future users. This work aims to understand the effects of this dynamic interaction. We propose a simple model where users with heterogeneous preferences arrive over time. Based on this model, we prove that naive estimators, i.e. those which ignore this feedback loop, are not consistent. We show that consistent estimators are efficient in the presence of myopic agents. Our results are validated using extensive simulations. } }
Endnote
%0 Conference Paper %T Human Interaction with Recommendation Systems %A Sven Schmit %A Carlos Riquelme %B Proceedings of the Twenty-First International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2018 %E Amos Storkey %E Fernando Perez-Cruz %F pmlr-v84-schmit18a %I PMLR %P 862--870 %U https://proceedings.mlr.press/v84/schmit18a.html %V 84 %X Many recommendation algorithms rely on user data to generate recommendations. However, these recommendations also affect the data obtained from future users. This work aims to understand the effects of this dynamic interaction. We propose a simple model where users with heterogeneous preferences arrive over time. Based on this model, we prove that naive estimators, i.e. those which ignore this feedback loop, are not consistent. We show that consistent estimators are efficient in the presence of myopic agents. Our results are validated using extensive simulations.
APA
Schmit, S. & Riquelme, C.. (2018). Human Interaction with Recommendation Systems. Proceedings of the Twenty-First International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 84:862-870 Available from https://proceedings.mlr.press/v84/schmit18a.html.

Related Material