Efficient Balanced Treatment Assignments for Experimentation

David Arbour, Drew Dimmery, Anup Rao
Proceedings of The 24th International Conference on Artificial Intelligence and Statistics, PMLR 130:3070-3078, 2021.

Abstract

In this work, we address the problem of balanced treatment assignment for experiments by considering an interpretation of the problem as optimization of a two-sample test between test and control units. Using this lens we provide an assignment algorithm that is optimal with respect to the minimum spanning tree test of Friedman and Rafsky [1979]. This assignment to treatment groups may be performed exactly in polynomial time and allows for the design of experiments explicitly targeting the individual treatment effect. We provide a probabilistic interpretation of this process in terms of the most probable element of designs drawn from a determinantal point process. We provide a novel formulation of estimation as transductive inference and show how the tree structures used in design can also be used in an adjustment estimator. We conclude with a simulation study demonstrating the improved efficacy of our method.

Cite this Paper

BibTeX
@InProceedings{pmlr-v130-arbour21a, title = { Efficient Balanced Treatment Assignments for Experimentation }, author = {Arbour, David and Dimmery, Drew and Rao, Anup}, booktitle = {Proceedings of The 24th International Conference on Artificial Intelligence and Statistics}, pages = {3070--3078}, year = {2021}, editor = {Banerjee, Arindam and Fukumizu, Kenji}, volume = {130}, series = {Proceedings of Machine Learning Research}, month = {13--15 Apr}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v130/arbour21a/arbour21a.pdf}, url = {https://proceedings.mlr.press/v130/arbour21a.html}, abstract = { In this work, we address the problem of balanced treatment assignment for experiments by considering an interpretation of the problem as optimization of a two-sample test between test and control units. Using this lens we provide an assignment algorithm that is optimal with respect to the minimum spanning tree test of Friedman and Rafsky [1979]. This assignment to treatment groups may be performed exactly in polynomial time and allows for the design of experiments explicitly targeting the individual treatment effect. We provide a probabilistic interpretation of this process in terms of the most probable element of designs drawn from a determinantal point process. We provide a novel formulation of estimation as transductive inference and show how the tree structures used in design can also be used in an adjustment estimator. We conclude with a simulation study demonstrating the improved efficacy of our method. } }
Endnote
%0 Conference Paper %T Efficient Balanced Treatment Assignments for Experimentation %A David Arbour %A Drew Dimmery %A Anup Rao %B Proceedings of The 24th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2021 %E Arindam Banerjee %E Kenji Fukumizu %F pmlr-v130-arbour21a %I PMLR %P 3070--3078 %U https://proceedings.mlr.press/v130/arbour21a.html %V 130 %X In this work, we address the problem of balanced treatment assignment for experiments by considering an interpretation of the problem as optimization of a two-sample test between test and control units. Using this lens we provide an assignment algorithm that is optimal with respect to the minimum spanning tree test of Friedman and Rafsky [1979]. This assignment to treatment groups may be performed exactly in polynomial time and allows for the design of experiments explicitly targeting the individual treatment effect. We provide a probabilistic interpretation of this process in terms of the most probable element of designs drawn from a determinantal point process. We provide a novel formulation of estimation as transductive inference and show how the tree structures used in design can also be used in an adjustment estimator. We conclude with a simulation study demonstrating the improved efficacy of our method.
APA
Arbour, D., Dimmery, D. & Rao, A.. (2021). Efficient Balanced Treatment Assignments for Experimentation . Proceedings of The 24th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 130:3070-3078 Available from https://proceedings.mlr.press/v130/arbour21a.html.

Related Material