Time Series Domain Adaptation Via Latent Invariant Causal Mechanism

Page view(s)
0
Checked on
Time Series Domain Adaptation Via Latent Invariant Causal Mechanism
Title:
Time Series Domain Adaptation Via Latent Invariant Causal Mechanism
Journal Title:
IEEE Transactions on Pattern Analysis and Machine Intelligence
Keywords:
Publication Date:
10 December 2025
Citation:
Cai, R., Huang, J., Yang, Z., Li, Z., Eldele, E., Wu, M., & Sun, F. (2025). Time Series Domain Adaptation Via Latent Invariant Causal Mechanism. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1–18. https://doi.org/10.1109/tpami.2025.3642245
Abstract:
Time series domain adaptation aims to transfer the complex temporal dependence from the labeled source domain to the unlabeled target domain. Recent advances leverage the stable causal mechanism over observed variables to model the domain-invariant temporal dependence. However, modeling precise causal structures in high-dimensional data, such as videos, remains challenging. Additionally, direct causal edges may not exist among observed variables (e.g., pixels). These limitations hinder the applicability of existing approaches to real-world scenarios. To address these challenges, we find that the high-dimension time series data are generated from the low-dimension latent variables, which motivates us to model the causal mechanisms of the temporal latent process. Based on this intuition, we propose a latent causal mechanism identification framework that guarantees the uniqueness of the reconstructed latent causal structures. Specifically, we first identify latent variables by utilizing sufficient changes in historical information. Moreover, by enforcing the sparsity of the relationships of latent variables, we can achieve identifiable latent causal structures. Built on the theoretical results, we develop the Latent Causality Alignment (LCA) model that leverages variational inference, which incorporates an intra-domain latent sparsity constraint for latent structure reconstruction and an inter-domain latent sparsity constraint for domain-invariant structure reconstruction. Experiment results on eight benchmarks show a general improvement in the domain-adaptive time series classification and forecasting tasks, highlighting the effectiveness of our method in real-world scenarios. Codes are available at https://github.com/DMIRLAB-Group/LCA.
License type:
Publisher Copyright
Funding Info:
This research / project is supported by the National Key R&D Program of China - N.A.
Grant Reference no. : 2021ZD0111501

This research / project is supported by the Natural Science Foundation of China - Excellent Young Scholars
Grant Reference no. : 62122022

This research / project is supported by the Natural Science Foundation of China - N.A.
Grant Reference no. : 61876043; 61976052

This research / project is supported by the Major Key Project of PCL - N.A.
Grant Reference no. : PCL2021A12
Description:
© 2025 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
ISSN:
0162-8828
2160-9292
1939-3539
Files uploaded:

File Size Format Action
zhenhui-zijian-lca-tpami-postprint.pdf 1.88 MB PDF Open