Learning Counterfactual Representations for Estimating Individual Dose-Response Curves

Learning Counterfactual Representations for Estimating Individual Dose-Response Curves. Schwab, P., Linhardt, L., Bauer, S., Buhmann, J., M., & Karlen, W. Proceedings of the AAAI Conference on Artificial Intelligence, 34(04):5612-5619, AAAI Press, 4, 2020.

Website doi abstract bibtex

Estimating what would be an individual's potential response to varying levels of exposure to a treatment is of high practical relevance for several important fields, such as healthcare, economics and public policy. However, existing methods for learning to estimate counterfactual outcomes from observational data are either focused on estimating average dose-response curves, or limited to settings with only two treatments that do not have an associated dosage parameter. Here, we present a novel machine-learning approach towards learning counterfactual representations for estimating individual dose-response curves for any number of treatments with continuous dosage parameters with neural networks. Building on the established potential outcomes framework, we introduce performance metrics, model selection criteria, model architectures, and open benchmarks for estimating individual dose-response curves. Our experiments show that the methods developed in this work set a new state-of-the-art in estimating individual dose-response.

@article{
 title = {Learning Counterfactual Representations for Estimating Individual Dose-Response Curves},
 type = {article},
 year = {2020},
 pages = {5612-5619},
 volume = {34},
 websites = {http://arxiv.org/abs/1902.00981,https://aaai.org/ojs/index.php/AAAI/article/view/6014},
 month = {4},
 publisher = {AAAI Press},
 day = {3},
 city = {New York},
 id = {76da0f6c-37d3-30a3-a25f-179ca2e70ba3},
 created = {2020-12-08T14:35:27.829Z},
 file_attached = {true},
 profile_id = {6d353feb-efe4-367e-84a2-0815eb9ca878},
 last_modified = {2022-09-04T18:12:03.032Z},
 read = {true},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Schwab2020},
 notes = {Acceptance rate: 0.20},
 folder_uuids = {f1f67efc-95a7-4f1a-b181-c3670c667a34,60555479-b7f0-45f5-aa97-a3920f93c426,4afa922c-d8d6-102e-ac9a-0024e85ead87,0801d9e0-d1ec-46e2-803d-c74946b43a02,d9198259-8733-497d-ab87-d2a9518e0d30,8bf4ed74-a467-49ad-8f7c-6ab98c981268},
 private_publication = {false},
 abstract = {Estimating what would be an individual's potential response to varying levels of exposure to a treatment is of high practical relevance for several important fields, such as healthcare, economics and public policy. However, existing methods for learning to estimate counterfactual outcomes from observational data are either focused on estimating average dose-response curves, or limited to settings with only two treatments that do not have an associated dosage parameter. Here, we present a novel machine-learning approach towards learning counterfactual representations for estimating individual dose-response curves for any number of treatments with continuous dosage parameters with neural networks. Building on the established potential outcomes framework, we introduce performance metrics, model selection criteria, model architectures, and open benchmarks for estimating individual dose-response curves. Our experiments show that the methods developed in this work set a new state-of-the-art in estimating individual dose-response.},
 bibtype = {article},
 author = {Schwab, Patrick and Linhardt, Lorenz and Bauer, Stefan and Buhmann, Joachim M. and Karlen, Walter},
 doi = {10.1609/aaai.v34i04.6014},
 journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
 number = {04}
}

Downloads: 0