Optimal Information Retention for Time-Series Explanations

Jinghang Yue, Jing Wang, Lu Zhang, Shuo Zhang, Da Li, Zhaoyang Ma, Youfang Lin
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:73851-73876, 2025.

Abstract

Explaining deep models for time-series data is crucial for identifying key patterns in sensitive domains, such as healthcare and finance. However, due to the lack of unified optimization criterion, existing explanation methods often suffer from redundancy and incompleteness, where irrelevant patterns are included or key patterns are missed in explanations. To address this challenge, we propose the Optimal Information Retention Principle, where conditional mutual information defines minimizing redundancy and maximizing completeness as optimization objectives. We then derive the corresponding objective function theoretically. As a practical framework, we introduce an explanation framework ORTE, learning a binary mask to eliminate redundant information while mining temporal patterns of explanations. We decouple the discrete mapping process to ensure the stability of gradient propagation, while employing contrastive learning to achieve precise filtering of explanatory patterns through the mask, thereby realizing a trade-off between low redundancy and high completeness. Extensive quantitative and qualitative experiments on synthetic and real-world datasets demonstrate that the proposed principle significantly improves the accuracy and completeness of explanations compared to baseline methods. The code is available at https://github.com/moon2yue/ORTE_public.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-yue25d, title = {Optimal Information Retention for Time-Series Explanations}, author = {Yue, Jinghang and Wang, Jing and Zhang, Lu and Zhang, Shuo and Li, Da and Ma, Zhaoyang and Lin, Youfang}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {73851--73876}, 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/yue25d/yue25d.pdf}, url = {https://proceedings.mlr.press/v267/yue25d.html}, abstract = {Explaining deep models for time-series data is crucial for identifying key patterns in sensitive domains, such as healthcare and finance. However, due to the lack of unified optimization criterion, existing explanation methods often suffer from redundancy and incompleteness, where irrelevant patterns are included or key patterns are missed in explanations. To address this challenge, we propose the Optimal Information Retention Principle, where conditional mutual information defines minimizing redundancy and maximizing completeness as optimization objectives. We then derive the corresponding objective function theoretically. As a practical framework, we introduce an explanation framework ORTE, learning a binary mask to eliminate redundant information while mining temporal patterns of explanations. We decouple the discrete mapping process to ensure the stability of gradient propagation, while employing contrastive learning to achieve precise filtering of explanatory patterns through the mask, thereby realizing a trade-off between low redundancy and high completeness. Extensive quantitative and qualitative experiments on synthetic and real-world datasets demonstrate that the proposed principle significantly improves the accuracy and completeness of explanations compared to baseline methods. The code is available at https://github.com/moon2yue/ORTE_public.} }
Endnote
%0 Conference Paper %T Optimal Information Retention for Time-Series Explanations %A Jinghang Yue %A Jing Wang %A Lu Zhang %A Shuo Zhang %A Da Li %A Zhaoyang Ma %A Youfang Lin %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-yue25d %I PMLR %P 73851--73876 %U https://proceedings.mlr.press/v267/yue25d.html %V 267 %X Explaining deep models for time-series data is crucial for identifying key patterns in sensitive domains, such as healthcare and finance. However, due to the lack of unified optimization criterion, existing explanation methods often suffer from redundancy and incompleteness, where irrelevant patterns are included or key patterns are missed in explanations. To address this challenge, we propose the Optimal Information Retention Principle, where conditional mutual information defines minimizing redundancy and maximizing completeness as optimization objectives. We then derive the corresponding objective function theoretically. As a practical framework, we introduce an explanation framework ORTE, learning a binary mask to eliminate redundant information while mining temporal patterns of explanations. We decouple the discrete mapping process to ensure the stability of gradient propagation, while employing contrastive learning to achieve precise filtering of explanatory patterns through the mask, thereby realizing a trade-off between low redundancy and high completeness. Extensive quantitative and qualitative experiments on synthetic and real-world datasets demonstrate that the proposed principle significantly improves the accuracy and completeness of explanations compared to baseline methods. The code is available at https://github.com/moon2yue/ORTE_public.
APA
Yue, J., Wang, J., Zhang, L., Zhang, S., Li, D., Ma, Z. & Lin, Y.. (2025). Optimal Information Retention for Time-Series Explanations. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:73851-73876 Available from https://proceedings.mlr.press/v267/yue25d.html.

Related Material