Graph Machine Learning based Doubly Robust Estimator for Network Causal Effects

Seyedeh Baharan Khatami, Harsh Parikh, Haowei Chen, Sudeepa Roy, Babak Salimi
Proceedings of The 28th International Conference on Artificial Intelligence and Statistics, PMLR 258:4366-4374, 2025.

Abstract

Estimating causal effects in social network data presents unique challenges due to the presence of spillover effects and network-induced confounding. While much of the existing literature addresses causal inference in social networks, many methods rely on strong assumptions about the form of network-induced confounding. These assumptions often fail to hold in high-dimensional networks, limiting the applicability of such approaches. To address this, we propose a novel methodology that integrates graph machine learning techniques with the double machine learning framework, facilitating accurate and efficient estimation of both direct and peer effects in a single observational social network. Our estimator achieves semiparametric efficiency under mild regularity conditions, enabling consistent uncertainty quantification. Through extensive simulations, we demonstrate the accuracy, robustness, and scalability of our method. Finally, we apply the proposed approach to examine the impact of Self-Help Group participation on financial risk tolerance, highlighting its practical relevance.

Cite this Paper

BibTeX
@InProceedings{pmlr-v258-khatami25a, title = {Graph Machine Learning based Doubly Robust Estimator for Network Causal Effects}, author = {Khatami, Seyedeh Baharan and Parikh, Harsh and Chen, Haowei and Roy, Sudeepa and Salimi, Babak}, booktitle = {Proceedings of The 28th International Conference on Artificial Intelligence and Statistics}, pages = {4366--4374}, year = {2025}, editor = {Li, Yingzhen and Mandt, Stephan and Agrawal, Shipra and Khan, Emtiyaz}, volume = {258}, series = {Proceedings of Machine Learning Research}, month = {03--05 May}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v258/main/assets/khatami25a/khatami25a.pdf}, url = {https://proceedings.mlr.press/v258/khatami25a.html}, abstract = {Estimating causal effects in social network data presents unique challenges due to the presence of spillover effects and network-induced confounding. While much of the existing literature addresses causal inference in social networks, many methods rely on strong assumptions about the form of network-induced confounding. These assumptions often fail to hold in high-dimensional networks, limiting the applicability of such approaches. To address this, we propose a novel methodology that integrates graph machine learning techniques with the double machine learning framework, facilitating accurate and efficient estimation of both direct and peer effects in a single observational social network. Our estimator achieves semiparametric efficiency under mild regularity conditions, enabling consistent uncertainty quantification. Through extensive simulations, we demonstrate the accuracy, robustness, and scalability of our method. Finally, we apply the proposed approach to examine the impact of Self-Help Group participation on financial risk tolerance, highlighting its practical relevance.} }
Endnote
%0 Conference Paper %T Graph Machine Learning based Doubly Robust Estimator for Network Causal Effects %A Seyedeh Baharan Khatami %A Harsh Parikh %A Haowei Chen %A Sudeepa Roy %A Babak Salimi %B Proceedings of The 28th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2025 %E Yingzhen Li %E Stephan Mandt %E Shipra Agrawal %E Emtiyaz Khan %F pmlr-v258-khatami25a %I PMLR %P 4366--4374 %U https://proceedings.mlr.press/v258/khatami25a.html %V 258 %X Estimating causal effects in social network data presents unique challenges due to the presence of spillover effects and network-induced confounding. While much of the existing literature addresses causal inference in social networks, many methods rely on strong assumptions about the form of network-induced confounding. These assumptions often fail to hold in high-dimensional networks, limiting the applicability of such approaches. To address this, we propose a novel methodology that integrates graph machine learning techniques with the double machine learning framework, facilitating accurate and efficient estimation of both direct and peer effects in a single observational social network. Our estimator achieves semiparametric efficiency under mild regularity conditions, enabling consistent uncertainty quantification. Through extensive simulations, we demonstrate the accuracy, robustness, and scalability of our method. Finally, we apply the proposed approach to examine the impact of Self-Help Group participation on financial risk tolerance, highlighting its practical relevance.
APA
Khatami, S.B., Parikh, H., Chen, H., Roy, S. & Salimi, B.. (2025). Graph Machine Learning based Doubly Robust Estimator for Network Causal Effects. Proceedings of The 28th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 258:4366-4374 Available from https://proceedings.mlr.press/v258/khatami25a.html.

Related Material