Hybrid$^2$ Neural ODE Causal Modeling and an Application to Glycemic Response

Bob Junyi Zou, Matthew E Levine, Dessi P. Zaharieva, Ramesh Johari, Emily Fox
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:62934-62963, 2024.

Abstract

Hybrid models composing mechanistic ODE-based dynamics with flexible and expressive neural network components have grown rapidly in popularity, especially in scientific domains where such ODE-based modeling offers important interpretability and validated causal grounding (e.g., for counterfactual reasoning). The incorporation of mechanistic models also provides inductive bias in standard blackbox modeling approaches, critical when learning from small datasets or partially observed, complex systems. Unfortunately, as the hybrid models become more flexible, the causal grounding provided by the mechanistic model can quickly be lost. We address this problem by leveraging another common source of domain knowledge: ranking of treatment effects for a set of interventions, even if the precise treatment effect is unknown. We encode this information in a causal loss that we combine with the standard predictive loss to arrive at a hybrid loss that biases our learning towards causally valid hybrid models. We demonstrate our ability to achieve a win-win, state-of-the-art predictive performance and causal validity, in the challenging task of modeling glucose dynamics post-exercise in individuals with type 1 diabetes.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-zou24b, title = {Hybrid$^2$ Neural {ODE} Causal Modeling and an Application to Glycemic Response}, author = {Zou, Bob Junyi and Levine, Matthew E and Zaharieva, Dessi P. and Johari, Ramesh and Fox, Emily}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {62934--62963}, year = {2024}, editor = {Salakhutdinov, Ruslan and Kolter, Zico and Heller, Katherine and Weller, Adrian and Oliver, Nuria and Scarlett, Jonathan and Berkenkamp, Felix}, volume = {235}, series = {Proceedings of Machine Learning Research}, month = {21--27 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v235/main/assets/zou24b/zou24b.pdf}, url = {https://proceedings.mlr.press/v235/zou24b.html}, abstract = {Hybrid models composing mechanistic ODE-based dynamics with flexible and expressive neural network components have grown rapidly in popularity, especially in scientific domains where such ODE-based modeling offers important interpretability and validated causal grounding (e.g., for counterfactual reasoning). The incorporation of mechanistic models also provides inductive bias in standard blackbox modeling approaches, critical when learning from small datasets or partially observed, complex systems. Unfortunately, as the hybrid models become more flexible, the causal grounding provided by the mechanistic model can quickly be lost. We address this problem by leveraging another common source of domain knowledge: ranking of treatment effects for a set of interventions, even if the precise treatment effect is unknown. We encode this information in a causal loss that we combine with the standard predictive loss to arrive at a hybrid loss that biases our learning towards causally valid hybrid models. We demonstrate our ability to achieve a win-win, state-of-the-art predictive performance and causal validity, in the challenging task of modeling glucose dynamics post-exercise in individuals with type 1 diabetes.} }
Endnote
%0 Conference Paper %T Hybrid$^2$ Neural ODE Causal Modeling and an Application to Glycemic Response %A Bob Junyi Zou %A Matthew E Levine %A Dessi P. Zaharieva %A Ramesh Johari %A Emily Fox %B Proceedings of the 41st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Ruslan Salakhutdinov %E Zico Kolter %E Katherine Heller %E Adrian Weller %E Nuria Oliver %E Jonathan Scarlett %E Felix Berkenkamp %F pmlr-v235-zou24b %I PMLR %P 62934--62963 %U https://proceedings.mlr.press/v235/zou24b.html %V 235 %X Hybrid models composing mechanistic ODE-based dynamics with flexible and expressive neural network components have grown rapidly in popularity, especially in scientific domains where such ODE-based modeling offers important interpretability and validated causal grounding (e.g., for counterfactual reasoning). The incorporation of mechanistic models also provides inductive bias in standard blackbox modeling approaches, critical when learning from small datasets or partially observed, complex systems. Unfortunately, as the hybrid models become more flexible, the causal grounding provided by the mechanistic model can quickly be lost. We address this problem by leveraging another common source of domain knowledge: ranking of treatment effects for a set of interventions, even if the precise treatment effect is unknown. We encode this information in a causal loss that we combine with the standard predictive loss to arrive at a hybrid loss that biases our learning towards causally valid hybrid models. We demonstrate our ability to achieve a win-win, state-of-the-art predictive performance and causal validity, in the challenging task of modeling glucose dynamics post-exercise in individuals with type 1 diabetes.
APA
Zou, B.J., Levine, M.E., Zaharieva, D.P., Johari, R. & Fox, E.. (2024). Hybrid$^2$ Neural ODE Causal Modeling and an Application to Glycemic Response. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:62934-62963 Available from https://proceedings.mlr.press/v235/zou24b.html.

Related Material