EconBase
← All papers

Policy learning for many outcomes of interest: Combining optimal policy trees with multi-objective Bayesian optimisation

Patrick Rehill, Nicholas Biddle

arXiv 13 Dec 2022 · Machine Learning · publishedComputational Economics (2024) · 1 citations (OpenAlex)

arXiv:2212.06312 · PDF · DOI · OpenAlex · Extracted main text

Abstract

Methods for learning optimal policies use causal machine learning models to create human-interpretable rules for making choices around the allocation of different policy interventions. However, in realistic policy-making contexts, decision-makers often care about trade-offs between outcomes, not just single-mindedly maximising utility for one outcome. This paper proposes an approach termed Multi-Objective Policy Learning (MOPoL) which combines optimal decision trees for policy learning with a multi-objective Bayesian optimisation approach to explore the trade-off between multiple outcomes. It does this by building a Pareto frontier of non-dominated models for different hyperparameter settings which govern outcome weighting. The key here is that a low-cost greedy tree can be an accurate proxy for the very computationally costly optimal tree for the purposes of making decisions which means models can be repeatedly fit to learn a Pareto frontier. The method is applied to a real-world case-study of non-price rationing of anti-malarial medication in Kenya.

Citation extraction

52
references
87
in-text mentions
52
distinct cited
0
self-citations
12,169
main-text words

appendix boundary found by appendix_titled_section at “Appendix A - Variables used in application” · 97% of the source is main text. Read the extracted text to check this.

Most heavily cited references

The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.

ReferenceIntensityMentionsSectionsMain text
1athey_policy_2020 APACrefauthors Athey, S. \ Wager, S. APACrefauthor… 2020091.00094100%
2sverdrup_policytree_2020 APACrefauthors Sverdrup, E. , Kanodia, A. ,… (2020) 2020061.00053100%
3zhou_offline_2023 APACrefauthors Zhou, Z. , Athey, S. \ Wager, S. AP… (2022) 2023011.00053100%
4athey_generalized_2019 APACrefauthors Athey, S. , Tibshirani, J. \ W… (2019) 2019040.92843100%
5daulton_parallel_2021 APACrefauthors Daulton, S. , Balandat, M. \ Ba… (2021) 2021100.81142100%
6bertsimas_optimal_2017 APACrefauthors Bertsimas, D. \ Dunn, J. APACr… (2017) 2017070.64422100%
7bertsimas_optimal_2019 APACrefauthors Bertsimas, D. , Dunn, J. \ Mun… (2018) 2019040.64422100%
8chernozhukov_doubledebiased_2018 APACrefauthors Chernozhukov, V. , C… (2018) 2018020.64422100%
9kunzel_metalearners_2019 APACrefauthors Künzel, S R. , Sekhon, J S.… (2019) 2019030.64422100%
10morales-hernandez_survey_2022 APACrefauthors Morales-Hernández, A. ,… (2022) 2022110.64422100%

Showing the top 10 of 52 scored citations.