Treatment-Response Models for Counterfactual Reasoning with Continuous-time, Continuous-valued Interventions

Hossein Soleimani, Adarsh Subbaswamy, Suchi Saria
Proceedings of the 33rd Conference on Uncertainty in Artificial Intelligence, PMLR R15:819-828, 2017.

Abstract

Treatment effects can be estimated from ob- servational data as the difference in poten- tial outcomes. In this paper, we address the challenge of estimating the potential outcome when treatment-dose levels can vary continu- ously over time. Further, the outcome variable may not be measured at a regular frequency. Our proposed solution represents the treatment response curves using linear time-invariant dynamical systems—this provides a flexible means for modeling response over time to highly variable dose curves. Moreover, for multivariate data, the proposed method: un- covers shared structure in treatment response and the baseline across multiple markers; and, flexibly models challenging correlation struc- ture both across and within signals over time. For this, we build upon the framework of multiple-output Gaussian Processes. On sim- ulated and a challenging clinical dataset, we show significant gains in accuracy over state- of-the-art models.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR15-soleimani17a, title = {Treatment-Response Models for Counterfactual Reasoning with Continuous-time, Continuous-valued Interventions}, author = {Soleimani, Hossein and Subbaswamy, Adarsh and Saria, Suchi}, booktitle = {Proceedings of the 33rd Conference on Uncertainty in Artificial Intelligence}, pages = {819--828}, year = {2017}, editor = {Elidan, Gal and Kersting, Kristian}, volume = {R15}, series = {Proceedings of Machine Learning Research}, month = {11--15 Aug}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/r15/main/assets/soleimani17a/soleimani17a.pdf}, url = {https://proceedings.mlr.press/r15/soleimani17a.html}, abstract = {Treatment effects can be estimated from ob- servational data as the difference in poten- tial outcomes. In this paper, we address the challenge of estimating the potential outcome when treatment-dose levels can vary continu- ously over time. Further, the outcome variable may not be measured at a regular frequency. Our proposed solution represents the treatment response curves using linear time-invariant dynamical systems—this provides a flexible means for modeling response over time to highly variable dose curves. Moreover, for multivariate data, the proposed method: un- covers shared structure in treatment response and the baseline across multiple markers; and, flexibly models challenging correlation struc- ture both across and within signals over time. For this, we build upon the framework of multiple-output Gaussian Processes. On sim- ulated and a challenging clinical dataset, we show significant gains in accuracy over state- of-the-art models.}, note = {Reissued by PMLR on 04 October 2026.} }
Endnote
%0 Conference Paper %T Treatment-Response Models for Counterfactual Reasoning with Continuous-time, Continuous-valued Interventions %A Hossein Soleimani %A Adarsh Subbaswamy %A Suchi Saria %B Proceedings of the 33rd Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2017 %E Gal Elidan %E Kristian Kersting %F pmlr-vR15-soleimani17a %I PMLR %P 819--828 %U https://proceedings.mlr.press/r15/soleimani17a.html %V R15 %X Treatment effects can be estimated from ob- servational data as the difference in poten- tial outcomes. In this paper, we address the challenge of estimating the potential outcome when treatment-dose levels can vary continu- ously over time. Further, the outcome variable may not be measured at a regular frequency. Our proposed solution represents the treatment response curves using linear time-invariant dynamical systems—this provides a flexible means for modeling response over time to highly variable dose curves. Moreover, for multivariate data, the proposed method: un- covers shared structure in treatment response and the baseline across multiple markers; and, flexibly models challenging correlation struc- ture both across and within signals over time. For this, we build upon the framework of multiple-output Gaussian Processes. On sim- ulated and a challenging clinical dataset, we show significant gains in accuracy over state- of-the-art models. %Z Reissued by PMLR on 04 October 2026.
APA
Soleimani, H., Subbaswamy, A. & Saria, S.. (2017). Treatment-Response Models for Counterfactual Reasoning with Continuous-time, Continuous-valued Interventions. Proceedings of the 33rd Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R15:819-828 Available from https://proceedings.mlr.press/r15/soleimani17a.html. Reissued by PMLR on 04 October 2026.

Related Material