Causal Discovery in Action: Learning Chain-Reaction Mechanisms from Interventions

Panayiotis Panayiotou, Özgür Şimşek
Proceedings of the Fifth Conference on Causal Learning and Reasoning, PMLR 323:1545-1571, 2026.

Abstract

Causal discovery is challenging in general dynamical systems because, without strong structural assumptions, the underlying causal graph may not be identifiable even from interventional data. However, many real-world systems exhibit directional, cascade-like structure, in which components activate sequentially and upstream failures suppress downstream effects. We study causal discovery in such chain-reaction systems and show that the causal structure is uniquely identifiable from blocking interventions that prevent individual components from activating. We propose a minimal estimator with finite-sample guarantees, achieving exponential error decay and logarithmic sample complexity. Experiments on synthetic models and diverse chain-reaction environments demonstrate reliable recovery from a few interventions, while observational heuristics fail in regimes with delayed or overlapping causal effects.

Cite this Paper

BibTeX
@InProceedings{pmlr-v323-panayiotou26a, title = {Causal Discovery in Action: Learning Chain-Reaction Mechanisms from Interventions}, author = {Panayiotou, Panayiotis and {\c{S}}im{\c{s}}ek, {\"O}zg{\"u}r}, booktitle = {Proceedings of the Fifth Conference on Causal Learning and Reasoning}, pages = {1545--1571}, year = {2026}, editor = {Mazaheri, Bijan and Hanson, Niels Richard}, volume = {323}, series = {Proceedings of Machine Learning Research}, month = {06--08 Apr}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v323/main/assets/panayiotou26a/panayiotou26a.pdf}, url = {https://proceedings.mlr.press/v323/panayiotou26a.html}, abstract = {Causal discovery is challenging in general dynamical systems because, without strong structural assumptions, the underlying causal graph may not be identifiable even from interventional data. However, many real-world systems exhibit directional, cascade-like structure, in which components activate sequentially and upstream failures suppress downstream effects. We study causal discovery in such chain-reaction systems and show that the causal structure is uniquely identifiable from blocking interventions that prevent individual components from activating. We propose a minimal estimator with finite-sample guarantees, achieving exponential error decay and logarithmic sample complexity. Experiments on synthetic models and diverse chain-reaction environments demonstrate reliable recovery from a few interventions, while observational heuristics fail in regimes with delayed or overlapping causal effects.} }
Endnote
%0 Conference Paper %T Causal Discovery in Action: Learning Chain-Reaction Mechanisms from Interventions %A Panayiotis Panayiotou %A Özgür Şimşek %B Proceedings of the Fifth Conference on Causal Learning and Reasoning %C Proceedings of Machine Learning Research %D 2026 %E Bijan Mazaheri %E Niels Richard Hanson %F pmlr-v323-panayiotou26a %I PMLR %P 1545--1571 %U https://proceedings.mlr.press/v323/panayiotou26a.html %V 323 %X Causal discovery is challenging in general dynamical systems because, without strong structural assumptions, the underlying causal graph may not be identifiable even from interventional data. However, many real-world systems exhibit directional, cascade-like structure, in which components activate sequentially and upstream failures suppress downstream effects. We study causal discovery in such chain-reaction systems and show that the causal structure is uniquely identifiable from blocking interventions that prevent individual components from activating. We propose a minimal estimator with finite-sample guarantees, achieving exponential error decay and logarithmic sample complexity. Experiments on synthetic models and diverse chain-reaction environments demonstrate reliable recovery from a few interventions, while observational heuristics fail in regimes with delayed or overlapping causal effects.
APA
Panayiotou, P. & Şimşek, Ö.. (2026). Causal Discovery in Action: Learning Chain-Reaction Mechanisms from Interventions. Proceedings of the Fifth Conference on Causal Learning and Reasoning, in Proceedings of Machine Learning Research 323:1545-1571 Available from https://proceedings.mlr.press/v323/panayiotou26a.html.

Related Material