Root Cause Analysis of Outliers in Unknown Cyclic Graphs

Daniela Schkoda, Dominik Janzing
Proceedings of The 29th International Conference on Artificial Intelligence and Statistics, PMLR 300:622-630, 2026.

Abstract

We study the propagation of outliers in cyclic causal graphs with linear structural equations, tracing them back to one or several "root cause" nodes. We show that it is possible to identify a short list of potential root causes provided that the perturbation is sufficiently strong and propagates according to the same structural equations as in the normal mode. This shortlist consists of the true root causes together with those of its parents lying on a cycle with the root cause. Notably, our method does not require prior knowledge of the causal graph and yields encouraging results on simulated data and real data from biology and cloud computing.

Cite this Paper

BibTeX
@InProceedings{pmlr-v300-schkoda26a, title = { Root Cause Analysis of Outliers in Unknown Cyclic Graphs }, author = {Schkoda, Daniela and Janzing, Dominik}, booktitle = {Proceedings of The 29th International Conference on Artificial Intelligence and Statistics}, pages = {622--630}, year = {2026}, editor = {Khan, Emtiyaz and Li, Yingzhen and Solin, Arno and Ramdas, Aaditya}, volume = {300}, series = {Proceedings of Machine Learning Research}, month = {02--05 May}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v300/main/assets/schkoda26a/schkoda26a.pdf}, url = {https://proceedings.mlr.press/v300/schkoda26a.html}, abstract = { We study the propagation of outliers in cyclic causal graphs with linear structural equations, tracing them back to one or several "root cause" nodes. We show that it is possible to identify a short list of potential root causes provided that the perturbation is sufficiently strong and propagates according to the same structural equations as in the normal mode. This shortlist consists of the true root causes together with those of its parents lying on a cycle with the root cause. Notably, our method does not require prior knowledge of the causal graph and yields encouraging results on simulated data and real data from biology and cloud computing. } }
Endnote
%0 Conference Paper %T Root Cause Analysis of Outliers in Unknown Cyclic Graphs %A Daniela Schkoda %A Dominik Janzing %B Proceedings of The 29th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2026 %E Emtiyaz Khan %E Yingzhen Li %E Arno Solin %E Aaditya Ramdas %F pmlr-v300-schkoda26a %I PMLR %P 622--630 %U https://proceedings.mlr.press/v300/schkoda26a.html %V 300 %X We study the propagation of outliers in cyclic causal graphs with linear structural equations, tracing them back to one or several "root cause" nodes. We show that it is possible to identify a short list of potential root causes provided that the perturbation is sufficiently strong and propagates according to the same structural equations as in the normal mode. This shortlist consists of the true root causes together with those of its parents lying on a cycle with the root cause. Notably, our method does not require prior knowledge of the causal graph and yields encouraging results on simulated data and real data from biology and cloud computing.
APA
Schkoda, D. & Janzing, D.. (2026). Root Cause Analysis of Outliers in Unknown Cyclic Graphs . Proceedings of The 29th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 300:622-630 Available from https://proceedings.mlr.press/v300/schkoda26a.html.

Related Material