FIC-TSC: Learning Time Series Classification with Fisher Information Constraint

Xiwen Chen, Wenhui Zhu, Peijie Qiu, Hao Wang, Huayu Li, Zihan Li, Yalin Wang, Aristeidis Sotiras, Abolfazl Razi
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:9888-9909, 2025.

Abstract

Analyzing time series data is crucial to a wide spectrum of applications, including economics, online marketplaces, and human healthcare. In particular, time series classification plays an indispensable role in segmenting different phases in stock markets, predicting customer behavior, and classifying worker actions and engagement levels. These aspects contribute significantly to the advancement of automated decision-making and system optimization in real-world applications. However, there is a large consensus that time series data often suffers from domain shifts between training and test sets, which dramatically degrades the classification performance. Despite the success of (reversible) instance normalization in handling the domain shifts for time series regression tasks, its performance in classification is unsatisfactory. In this paper, we propose $\textit{FIC-TSC}$, a training framework for time series classification that leverages Fisher information as the constraint. We theoretically and empirically show this is an efficient and effective solution to guide the model converges toward flatter minima, which enhances its generalizability to distribution shifts. We rigorously evaluate our method on 30 UEA multivariate and 85 UCR univariate datasets. Our empirical results demonstrate the superiority of the proposed method over 14 recent state-of-the-art methods.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-chen25cq, title = {{FIC}-{TSC}: Learning Time Series Classification with {F}isher Information Constraint}, author = {Chen, Xiwen and Zhu, Wenhui and Qiu, Peijie and Wang, Hao and Li, Huayu and Li, Zihan and Wang, Yalin and Sotiras, Aristeidis and Razi, Abolfazl}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {9888--9909}, 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/chen25cq/chen25cq.pdf}, url = {https://proceedings.mlr.press/v267/chen25cq.html}, abstract = {Analyzing time series data is crucial to a wide spectrum of applications, including economics, online marketplaces, and human healthcare. In particular, time series classification plays an indispensable role in segmenting different phases in stock markets, predicting customer behavior, and classifying worker actions and engagement levels. These aspects contribute significantly to the advancement of automated decision-making and system optimization in real-world applications. However, there is a large consensus that time series data often suffers from domain shifts between training and test sets, which dramatically degrades the classification performance. Despite the success of (reversible) instance normalization in handling the domain shifts for time series regression tasks, its performance in classification is unsatisfactory. In this paper, we propose $\textit{FIC-TSC}$, a training framework for time series classification that leverages Fisher information as the constraint. We theoretically and empirically show this is an efficient and effective solution to guide the model converges toward flatter minima, which enhances its generalizability to distribution shifts. We rigorously evaluate our method on 30 UEA multivariate and 85 UCR univariate datasets. Our empirical results demonstrate the superiority of the proposed method over 14 recent state-of-the-art methods.} }
Endnote
%0 Conference Paper %T FIC-TSC: Learning Time Series Classification with Fisher Information Constraint %A Xiwen Chen %A Wenhui Zhu %A Peijie Qiu %A Hao Wang %A Huayu Li %A Zihan Li %A Yalin Wang %A Aristeidis Sotiras %A Abolfazl Razi %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-chen25cq %I PMLR %P 9888--9909 %U https://proceedings.mlr.press/v267/chen25cq.html %V 267 %X Analyzing time series data is crucial to a wide spectrum of applications, including economics, online marketplaces, and human healthcare. In particular, time series classification plays an indispensable role in segmenting different phases in stock markets, predicting customer behavior, and classifying worker actions and engagement levels. These aspects contribute significantly to the advancement of automated decision-making and system optimization in real-world applications. However, there is a large consensus that time series data often suffers from domain shifts between training and test sets, which dramatically degrades the classification performance. Despite the success of (reversible) instance normalization in handling the domain shifts for time series regression tasks, its performance in classification is unsatisfactory. In this paper, we propose $\textit{FIC-TSC}$, a training framework for time series classification that leverages Fisher information as the constraint. We theoretically and empirically show this is an efficient and effective solution to guide the model converges toward flatter minima, which enhances its generalizability to distribution shifts. We rigorously evaluate our method on 30 UEA multivariate and 85 UCR univariate datasets. Our empirical results demonstrate the superiority of the proposed method over 14 recent state-of-the-art methods.
APA
Chen, X., Zhu, W., Qiu, P., Wang, H., Li, H., Li, Z., Wang, Y., Sotiras, A. & Razi, A.. (2025). FIC-TSC: Learning Time Series Classification with Fisher Information Constraint. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:9888-9909 Available from https://proceedings.mlr.press/v267/chen25cq.html.

Related Material