Root Cause Identification for Collective Anomalies in Time Series given an Acyclic Summary Causal Graph with Loops

Charles K. Assaad, Imad Ez-Zejjari, Lei Zan
Proceedings of The 26th International Conference on Artificial Intelligence and Statistics, PMLR 206:8395-8404, 2023.

Abstract

This paper presents an approach for identifying the root causes of collective anomalies given observational time series and an acyclic summary causal graph which depicts an abstraction of causal relations present in a dynamic system at its normal regime. The paper first shows how the problem of root cause identification can be divided into many independent subproblems by grouping related anomalies using d-separation. Further, it shows how, under this setting, some root causes can be found directly from the graph and from the time of appearance of anomalies. Finally, it shows, how the rest of the root causes can be found by comparing direct causal effects in the normal and in the anomalous regime. To this end, temporal adaptations of the back-door and the single-door criterions are introduced. Extensive experiments conducted on both simulated and real-world datasets demonstrate the effectiveness of the proposed method.

Cite this Paper


BibTeX
@InProceedings{pmlr-v206-assaad23a, title = {Root Cause Identification for Collective Anomalies in Time Series given an Acyclic Summary Causal Graph with Loops}, author = {Assaad, Charles K. and Ez-Zejjari, Imad and Zan, Lei}, booktitle = {Proceedings of The 26th International Conference on Artificial Intelligence and Statistics}, pages = {8395--8404}, year = {2023}, editor = {Ruiz, Francisco and Dy, Jennifer and van de Meent, Jan-Willem}, volume = {206}, series = {Proceedings of Machine Learning Research}, month = {25--27 Apr}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v206/assaad23a/assaad23a.pdf}, url = {https://proceedings.mlr.press/v206/assaad23a.html}, abstract = {This paper presents an approach for identifying the root causes of collective anomalies given observational time series and an acyclic summary causal graph which depicts an abstraction of causal relations present in a dynamic system at its normal regime. The paper first shows how the problem of root cause identification can be divided into many independent subproblems by grouping related anomalies using d-separation. Further, it shows how, under this setting, some root causes can be found directly from the graph and from the time of appearance of anomalies. Finally, it shows, how the rest of the root causes can be found by comparing direct causal effects in the normal and in the anomalous regime. To this end, temporal adaptations of the back-door and the single-door criterions are introduced. Extensive experiments conducted on both simulated and real-world datasets demonstrate the effectiveness of the proposed method.} }
Endnote
%0 Conference Paper %T Root Cause Identification for Collective Anomalies in Time Series given an Acyclic Summary Causal Graph with Loops %A Charles K. Assaad %A Imad Ez-Zejjari %A Lei Zan %B Proceedings of The 26th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2023 %E Francisco Ruiz %E Jennifer Dy %E Jan-Willem van de Meent %F pmlr-v206-assaad23a %I PMLR %P 8395--8404 %U https://proceedings.mlr.press/v206/assaad23a.html %V 206 %X This paper presents an approach for identifying the root causes of collective anomalies given observational time series and an acyclic summary causal graph which depicts an abstraction of causal relations present in a dynamic system at its normal regime. The paper first shows how the problem of root cause identification can be divided into many independent subproblems by grouping related anomalies using d-separation. Further, it shows how, under this setting, some root causes can be found directly from the graph and from the time of appearance of anomalies. Finally, it shows, how the rest of the root causes can be found by comparing direct causal effects in the normal and in the anomalous regime. To this end, temporal adaptations of the back-door and the single-door criterions are introduced. Extensive experiments conducted on both simulated and real-world datasets demonstrate the effectiveness of the proposed method.
APA
Assaad, C.K., Ez-Zejjari, I. & Zan, L.. (2023). Root Cause Identification for Collective Anomalies in Time Series given an Acyclic Summary Causal Graph with Loops. Proceedings of The 26th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 206:8395-8404 Available from https://proceedings.mlr.press/v206/assaad23a.html.

Related Material