Arrow: Accelerator for Time Series Causal Discovery with Time Weaving

Yuanyuan Yao, Yuan Dong, Lu Chen, Kun Kuang, Ziquan Fang, Cheng Long, Yunjun Gao, Tianyi Li
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:71668-71686, 2025.

Abstract

Current causal discovery methods for time series data can effectively address a variety of scenarios; however, they remain constrained by inefficiencies. The significant inefficiencies arise primarily from the high computational costs associated with binning, the uncertainty in selecting appropriate time lags, and the extensive sets of candidate variables. To achieve both high efficiency and effectiveness of causal discovery, we introduce an accelerator termed ARROW. It incorporates an innovative concept termed “Time Weaving” that efficiently encodes time series data to well capture the dynamic trends, thereby mitigating computational complexity. We also propose a novel time lag discovery strategy utilizing XOR operations, which derives a theorem to obtain the optimal time lag and significantly enhances the efficiency using XOR operations. To optimize the search space for causal relationships, we design an efficient pruning strategy that intelligently identifies the most relevant candidate variables, enhancing the efficiency and accuracy of causal discovery. We applied ARROW to four different types of time series causal discovery algorithms and evaluated it on 25 synthetic and real-world datasets. The results demonstrate that, compared to the original algorithms, ARROW achieves up to 153x speedup while achieving higher accuracy in most cases.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-yao25a, title = {Arrow: Accelerator for Time Series Causal Discovery with Time Weaving}, author = {Yao, Yuanyuan and Dong, Yuan and Chen, Lu and Kuang, Kun and Fang, Ziquan and Long, Cheng and Gao, Yunjun and Li, Tianyi}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {71668--71686}, 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/yao25a/yao25a.pdf}, url = {https://proceedings.mlr.press/v267/yao25a.html}, abstract = {Current causal discovery methods for time series data can effectively address a variety of scenarios; however, they remain constrained by inefficiencies. The significant inefficiencies arise primarily from the high computational costs associated with binning, the uncertainty in selecting appropriate time lags, and the extensive sets of candidate variables. To achieve both high efficiency and effectiveness of causal discovery, we introduce an accelerator termed ARROW. It incorporates an innovative concept termed “Time Weaving” that efficiently encodes time series data to well capture the dynamic trends, thereby mitigating computational complexity. We also propose a novel time lag discovery strategy utilizing XOR operations, which derives a theorem to obtain the optimal time lag and significantly enhances the efficiency using XOR operations. To optimize the search space for causal relationships, we design an efficient pruning strategy that intelligently identifies the most relevant candidate variables, enhancing the efficiency and accuracy of causal discovery. We applied ARROW to four different types of time series causal discovery algorithms and evaluated it on 25 synthetic and real-world datasets. The results demonstrate that, compared to the original algorithms, ARROW achieves up to 153x speedup while achieving higher accuracy in most cases.} }
Endnote
%0 Conference Paper %T Arrow: Accelerator for Time Series Causal Discovery with Time Weaving %A Yuanyuan Yao %A Yuan Dong %A Lu Chen %A Kun Kuang %A Ziquan Fang %A Cheng Long %A Yunjun Gao %A Tianyi Li %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-yao25a %I PMLR %P 71668--71686 %U https://proceedings.mlr.press/v267/yao25a.html %V 267 %X Current causal discovery methods for time series data can effectively address a variety of scenarios; however, they remain constrained by inefficiencies. The significant inefficiencies arise primarily from the high computational costs associated with binning, the uncertainty in selecting appropriate time lags, and the extensive sets of candidate variables. To achieve both high efficiency and effectiveness of causal discovery, we introduce an accelerator termed ARROW. It incorporates an innovative concept termed “Time Weaving” that efficiently encodes time series data to well capture the dynamic trends, thereby mitigating computational complexity. We also propose a novel time lag discovery strategy utilizing XOR operations, which derives a theorem to obtain the optimal time lag and significantly enhances the efficiency using XOR operations. To optimize the search space for causal relationships, we design an efficient pruning strategy that intelligently identifies the most relevant candidate variables, enhancing the efficiency and accuracy of causal discovery. We applied ARROW to four different types of time series causal discovery algorithms and evaluated it on 25 synthetic and real-world datasets. The results demonstrate that, compared to the original algorithms, ARROW achieves up to 153x speedup while achieving higher accuracy in most cases.
APA
Yao, Y., Dong, Y., Chen, L., Kuang, K., Fang, Z., Long, C., Gao, Y. & Li, T.. (2025). Arrow: Accelerator for Time Series Causal Discovery with Time Weaving. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:71668-71686 Available from https://proceedings.mlr.press/v267/yao25a.html.

Related Material