Causality-Aware Contrastive Learning for Robust Multivariate Time-Series Anomaly Detection

Hyungi Kim, Jisoo Mok, Dongjun Lee, Jaihyun Lew, Sungjae Kim, Sungroh Yoon
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:30591-30608, 2025.

Abstract

Utilizing the complex inter-variable causal relationships within multivariate time-series provides a promising avenue toward more robust and reliable multivariate time-series anomaly detection (MTSAD) but remains an underexplored area of research. This paper proposes Causality-Aware contrastive learning for RObust multivariate Time-Series (CAROTS), a novel MTSAD pipeline that incorporates the notion of causality into contrastive learning. CAROTS employs two data augmentors to obtain causality-preserving and -disturbing samples that serve as a wide range of normal variations and synthetic anomalies, respectively. With causality-preserving and -disturbing samples as positives and negatives, CAROTS performs contrastive learning to train an encoder whose latent space separates normal and abnormal samples based on causality. Moreover, CAROTS introduces a similarity-filtered one-class contrastive loss that encourages the contrastive learning process to gradually incorporate more semantically diverse samples with common causal relationships. Extensive experiments on five real-world and two synthetic datasets validate that the integration of causal relationships endows CAROTS with improved MTSAD capabilities. The code is available at https://github.com/kimanki/CAROTS.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-kim25aa, title = {Causality-Aware Contrastive Learning for Robust Multivariate Time-Series Anomaly Detection}, author = {Kim, Hyungi and Mok, Jisoo and Lee, Dongjun and Lew, Jaihyun and Kim, Sungjae and Yoon, Sungroh}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {30591--30608}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/kim25aa/kim25aa.pdf}, url = {https://proceedings.mlr.press/v267/kim25aa.html}, abstract = {Utilizing the complex inter-variable causal relationships within multivariate time-series provides a promising avenue toward more robust and reliable multivariate time-series anomaly detection (MTSAD) but remains an underexplored area of research. This paper proposes Causality-Aware contrastive learning for RObust multivariate Time-Series (CAROTS), a novel MTSAD pipeline that incorporates the notion of causality into contrastive learning. CAROTS employs two data augmentors to obtain causality-preserving and -disturbing samples that serve as a wide range of normal variations and synthetic anomalies, respectively. With causality-preserving and -disturbing samples as positives and negatives, CAROTS performs contrastive learning to train an encoder whose latent space separates normal and abnormal samples based on causality. Moreover, CAROTS introduces a similarity-filtered one-class contrastive loss that encourages the contrastive learning process to gradually incorporate more semantically diverse samples with common causal relationships. Extensive experiments on five real-world and two synthetic datasets validate that the integration of causal relationships endows CAROTS with improved MTSAD capabilities. The code is available at https://github.com/kimanki/CAROTS.} }
Endnote
%0 Conference Paper %T Causality-Aware Contrastive Learning for Robust Multivariate Time-Series Anomaly Detection %A Hyungi Kim %A Jisoo Mok %A Dongjun Lee %A Jaihyun Lew %A Sungjae Kim %A Sungroh Yoon %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-kim25aa %I PMLR %P 30591--30608 %U https://proceedings.mlr.press/v267/kim25aa.html %V 267 %X Utilizing the complex inter-variable causal relationships within multivariate time-series provides a promising avenue toward more robust and reliable multivariate time-series anomaly detection (MTSAD) but remains an underexplored area of research. This paper proposes Causality-Aware contrastive learning for RObust multivariate Time-Series (CAROTS), a novel MTSAD pipeline that incorporates the notion of causality into contrastive learning. CAROTS employs two data augmentors to obtain causality-preserving and -disturbing samples that serve as a wide range of normal variations and synthetic anomalies, respectively. With causality-preserving and -disturbing samples as positives and negatives, CAROTS performs contrastive learning to train an encoder whose latent space separates normal and abnormal samples based on causality. Moreover, CAROTS introduces a similarity-filtered one-class contrastive loss that encourages the contrastive learning process to gradually incorporate more semantically diverse samples with common causal relationships. Extensive experiments on five real-world and two synthetic datasets validate that the integration of causal relationships endows CAROTS with improved MTSAD capabilities. The code is available at https://github.com/kimanki/CAROTS.
APA
Kim, H., Mok, J., Lee, D., Lew, J., Kim, S. & Yoon, S.. (2025). Causality-Aware Contrastive Learning for Robust Multivariate Time-Series Anomaly Detection. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:30591-30608 Available from https://proceedings.mlr.press/v267/kim25aa.html.

Related Material